Cortical synergistic information transmission is associated with obesity-related differences in working memory through different interaction patterns

Background Obesity has been increasingly linked to differences in working memory (WM), yet the neurophysiological mechanisms linking metabolic state to cognitive function remain unclear. Here, we investigated whether alterations in cortical information transmission are associated with obesity-related differences in working memory. Methods Using resting-state fMRI data from the Human Connectome Project, we applied an information-theoretic framework to quantify synergistic and redundant connectivity among key brain regions supporting working memory. We examined whether network-level and edge-level connectivity features were associated with BMI and working memory performance and further assessed their moderating relationships with the BMI–WM association, together with exploratory indirect-effect analyses. Results Higher BMI was associated with poorer WM performance (β = − 0.149, 95% CI [− 0.256, − 0.041]) and reduced efficiency of synergistic information processing within large-scale brain networks (β = − 0.134, 95% CI [− 0.241, − 0.026]). Reduced synergistic connectivity between the anterior insula and dorsomedial prefrontal cortex ( p = 0.032, Cohen's d = 0.308) was associated with a weaker negative relationship between BMI and working memory. Different connectivity measures also revealed different patterns of association with BMI and WM performance. Conclusions These findings suggest that BMI is associated with working memory through different patterns of network-level information processing, including reduced efficiency of synergistic information integration and connectivity patterns that influence the relationship between BMI and WM. Together, our results highlight synergistic information processing as a potential marker of network-level information organization and provide a framework for investigating links between metabolic status and cognitive function.