Brain Networks Associated With COVID-19 Risk: Data From 3,662 Participants

Chadi G. AbdallahmedRxiv 2021 · DOI

Abstract

While much research has focused on the neurological consequences of COVID-19, this study investigates whether pre-existing brain architecture influences the risk of infection. Using data from the UK Biobank, researchers identified specific functional brain network patterns—"connectivity fingerprints"—that correlate with a higher likelihood of testing positive for the virus. The study found that individuals with reduced connectivity in top-down "control" networks and increased internal connectivity in "instinctive" networks were more likely to test positive, suggesting that neural markers for risk-taking behavior may predate and predict infection risk.

Introduction

Does the way your brain is wired influence your likelihood of getting sick during a pandemic? While biological susceptibility is often discussed in terms of genetics or immune health, behavioral traits—such as impulsivity, risk-taking, and adherence to public health guidelines—are significant drivers of viral exposure. Since these traits have known neural correlates, this research sought to identify specific functional brain network patterns (connectomes) that serve as predisposing biomarkers for COVID-19 risk. By analyzing brain scans taken years before the pandemic, researchers aimed to disentangle pre-existing brain signatures from the neurological consequences of the disease.

Methods

The study utilized a "network-informed" machine learning approach called Network-Restricted Strength Predictive Modeling (NRS-PM) to analyze data from the UK Biobank Resource.

Participants and Data

The cohort consisted of 3,662 individuals (average age 63 at the time of scanning) who underwent structural and functional MRI scans. Crucially, the imaging data were acquired an average of three years prior to COVID-19 testing (scans from ~2017; testing up to February 2021). This longitudinal design ensures that the observed brain patterns were present before any potential exposure to the virus.

Connectivity Mapping

Using resting-state and task-based fMRI, the researchers mapped the "functional connectome." They employed the Akiki-Abdallah (AA) atlas to divide the brain into 7 canonical networks (including the Central Executive, Default Mode, and Ventral Salience networks) and further into 24 and 50 sub-modules. The NRS-PM framework allowed the team to identify "Connectivity Fingerprints" (CFPs) and "Nodal Fingerprints" (NFPs), translating predictive features back into neurobiological space to identify which specific network connections drove the risk predictions.

Key Results

The predictive models successfully identified six significant fingerprints associated with COVID-19 status (all p < 0.005). The findings highlight a fundamental shift in brain organization among those who later tested positive:

  • Integration vs. Segregation: COVID-positive individuals showed a pattern of increased neural segregation. This was characterized by higher internal connectivity within specific modules but lower "integration" (communication) across different brain networks.
  • The "Control vs. Instinct" Imbalance: Infection risk was strongly associated with reduced connectivity between "top-down" attention/executive networks (Dorsal Salience and Central Executive) and "introspective/instinctive" networks (Default Mode and Ventral Salience).
  • Increased Internal Signaling: Higher infection rates correlated with more intense internal signaling within the Default Mode, Ventral Salience, subcortical, and sensorimotor networks.
  • Regional Strength: Positive status was specifically associated with increased overall functional connectivity strength in the insula and medial frontal areas—regions heavily involved in emotion and saliency processing.

Discussion

The identified patterns—specifically reduced frontoparietal connectivity and increased insular connectivity—mirror brain signatures often observed in addiction and impulse control disorders. This suggests the model is detecting a "risk-taking phenotype" that leads to higher viral exposure through behavioral choices.

Because the scans predated the pandemic, these findings rule out the possibility that the observed network changes were caused by the virus itself. Instead, they point toward a pre-existing neural vulnerability. While the study is associative and lacks direct behavioral measures (such as mask-wearing frequency), it provides a novel framework for understanding how brain architecture contributes to public health outcomes. The authors suggest that these risk networks could eventually serve as targets for clinical interventions, such as neuromodulation, aimed at improving impulse control in vulnerable populations.

Summary updated

Abstract

Abstract Background Our behavioral traits, and subsequent actions, could affect the risk of exposure to the coronavirus disease of 2019 (COVID-19). The current study aimed to determine whether unique brain networks are associated with the COVID-19 infection risk. Methods This research was conducted using the UK Biobank Resource. Functional magnetic resonance imaging scans in a cohort of general population (n=3,662) were used to compute the whole-brain functional connectomes. A network-informed machine learning approach was used to identify connectome and nodal fingerprints that are associated with positive COVID-19 status during the pandemic up to February 4 th , 2021. Results The predictive models successfully identified 6 fingerprints that were associated with COVID-19 positive, compared to negative status (all p values &lt; 0.005). Overall, lower integration across the brain modules and increased segregation, as reflected by internal within module connectivity, were associated with higher infection rates. More specifically, COVID-19 positive status was associated with 1) reduced connectivity between the central executive and ventral salience, as well as between the dorsal salience and default mode networks; 2) increased internal connectivity within the default mode, ventral salience, subcortical and sensorimotor networks; and 3) increased connectivity between the ventral salience, subcortical and sensorimotor networks. Conclusion Individuals are at increased risk of COVID-19 infections if their brain connectome is consistent with reduced connectivity in the top-down attention and executive networks, along with increased internal connectivity in the introspective and instinctive networks. These identified risk networks could be investigated as target for treatment of illnesses with impulse control deficits.
BibTeX
@misc{abdallah2021brain,
  author = {Chadi G. Abdallah},
  title = {{Brain Networks Associated With COVID-19 Risk: Data From 3,662 Participants}},
  year = {2021},
  doi = {10.1101/2021.04.01.21254709},
  howpublished = {medRxiv},
}
Figure