Characterizing the Relationship Between Functional Network Dynamics and the Body Mass Index

Why It Matters

The findings reveal a neural signature of obesity that ties higher BMI to less dynamic visual‑network activity, offering potential biomarkers for early detection and targets for interventions aimed at improving brain flexibility.

Key Takeaways

• Higher BMI increases time spent in visual‑network dominant FC state
• BMI linked to higher fractional window and dwell time of that state
• BMI inversely related to variability of nodal efficiency in visual regions
• Visual network rigidity may underlie heightened food cue sensitivity in obesity
• Large HCP sample (n=776) strengthens confidence in dynamic connectivity results

Pulse Analysis

Obesity remains a global health crisis, with the World Obesity Atlas projecting over a billion adults will be classified as obese by 2030. Researchers rely on body mass index (BMI) as a convenient proxy for adiposity, but the metric’s utility extends into neuroscience, where it serves as a bridge between metabolic health and brain function. While static resting‑state functional connectivity has identified altered interactions among default‑mode, salience, and executive networks in higher‑BMI individuals, these approaches assume a stationary brain and miss the rapid fluctuations that characterize neural communication. Dynamic functional connectivity (dFC) offers a more nuanced view, capturing moment‑to‑moment changes that may underlie behavioral traits such as food cue reactivity.

In the new analysis, investigators leveraged the high‑quality resting‑state fMRI data of 776 participants from the Human Connectome Project. Using a sliding‑window technique and k‑means clustering, they extracted four recurring whole‑brain connectivity states. One state, distinguished by strong intra‑ and inter‑network coupling within the visual network (VN), showed a modest but significant positive correlation between BMI and both fractional window and mean dwell time. Parallel graph‑theoretic calculations revealed that higher BMI corresponded to lower temporal variability of nodal efficiency and local efficiency in VN‑related independent components, indicating a more rigid visual processing architecture. These converging lines of evidence point to a VN‑centric pattern of reduced neural flexibility in individuals with elevated BMI.

The implications are twofold. First, the VN‑dominant state and its associated topological stability could serve as early neuroimaging biomarkers for obesity risk, complementing metabolic and behavioral assessments. Second, interventions that promote neural flexibility—such as cognitive training, mindfulness, or neuromodulation—might mitigate the heightened visual salience of food cues that drives overeating. Future work should test longitudinal designs to untangle causality, expand the age range beyond young adults, and integrate cognitive performance metrics to clarify how visual‑network rigidity translates into real‑world eating behaviors. Despite modest effect sizes, the study underscores the value of dynamic brain metrics in decoding the complex neurobiology of obesity.