Human intelligence has always been a fascinating subject for scientists. Since the inception of Spearman’s general intelligence in the early 1900s, there has been significant progress towards characterizing different aspects of intelligence and its relationship with structural and functional features of the brain. In recent years, the invention of sophisticated brain imaging devices using Diffusion-Weighted Imaging (DWI) and functional Magnetic Resonance Imaging (fMRI) has allowed researchers to test hypotheses about neural correlates of intelligence in humans.This review summarizes recent findings on the associations of human intelligence with neuroimaging data. To this end, first, we review the literature that has related brain morphometry to intelligence. Next, we elaborate on the applications of DWI and resting-state fMRI on the investigation of intelligence. Then, we provide a survey of literature that has used multimodal DWI-fMRI to shed light on intelligence. Finally, we discuss the state-of-the-art of individualized prediction of intelligence from neuroimaging data and point out future strategies. Future studies hold promising outcomes for machine learning-based predictive frameworks using neuroimaging features to estimate human intelligence.

Recognition of a brain region’s interaction is an important field in neuroscience. Most studies use the Pearson correlation to find the interaction between the regions. According to the experimental evidence, there is a nonlinear dependence between the activities of different brain regions that is ignored by Pearson correlation as a linear measure. Typically, the average activity of each region is used as input because it is a univariate measure. This dimensional reduction, i.e., averaging, leads to a loss of spatial information across voxels within the region. In this study, we propose using an information-theoretic measure, multivariate mutual information (mvMI), as a nonlinear dependence to find the interaction between regions. This measure, which has been recently proposed, simplifies the mutual information calculation complexity using the Gaussian copula. Using simulated data, we show that the using this measure overcomes the mentioned limitations. Additionally using the real resting-state fMRI data, we compare the level of significance and randomness of graphs constructed using different methods. Our results indicate that the proposed method estimates the functional connectivity more significantly and leads to a smaller number of random connections than the common measure, Pearson correlation. Moreover, we find that the similarity of the estimated functional networks of the individuals is higher when the proposed method is used.

The brain’s capacity for integration arises from both its structural wiring and energetically demanding electrochemical signaling. Yet current connectome analyses treat network nodes as functionally homogeneous, ignoring that neural communication is constrained by metabolic cost. Here, we introduce a metabolism-weighted connectome, a fully weighted brain graph in which both connections and the metabolic activity of each node describe the network’s capacity for integration. Using three datasets of simultaneous fMRI and [ ¹⁸ F]Fluorodeoxyglucose positron emission tomography acquisitions, we define metabolism-weighted centrality (MwC), a biologically grounded index of each region’s signaling dominance that integrates functional connectivity with local energy metabolism. MwC provides a more accurate representation of cortical activity flow than classical edge-based metrics and reveals that metabolically active hubs align with higher-order cognitive networks. Transcriptomic and synaptic imaging data demonstrate that these hubs exhibit increased synaptic energy turnover, linking activity-driven centrality to the molecular architecture of signaling. Notably, the same high-MwC regions show greater susceptibility to neurodegenerative pathology, suggesting that lifelong metabolic demand influences both integrative function and disease vulnerability. By linking neuronal metabolism to network organization, our framework bridges cellular energetics and system-level computation, opening broad avenues for interpreting brain vulnerability and performance.

The brain’s capacity for integration arises from both its structural wiring and energetically demanding electrochemical signaling. Yet current connectome analyses treat network nodes as functionally homogeneous, ignoring that neural communication is constrained by metabolic cost. Here we introduce a metabolism-weighted connectome, a fully weighted brain graph in which both connections and the metabolic activity of each node describe the network’s capacity for integration. Using three datasets of simultaneous fMRI and FDG-PET acquisitions, we define metabolism-weighted centrality (MwC), a biologically grounded index of each region’s signaling dominance that integrates functional connectivity (FC) with local energy metabolism. MwC provides a more accurate representation of cortical activity flow than classical edge-based metrics and reveals that metabolically active hubs align with higher-order cognitive networks. Transcriptomic and synaptic imaging data demonstrate that these hubs exhibit increased synaptic energy turnover, linking activity-driven centrality to the molecular architecture of signaling. Notably, the same high-MwC regions show greater susceptibility to neurodegenerative pathology, suggesting that lifelong metabolic demand influences both integrative function and disease vulnerability. By linking neuronal metabolism to network organization, our framework bridges cellular energetics and system-level computation, opening new avenues for interpreting brain vulnerability and performance. Significance Statement: The various regions of the brain exhibit unequal energy consumption. Specifically, areas associated with cognitive functions such as memory and attention are metabolic hotspots of intense neural activity. Yet standard connectivity analyses ignore this energetic dimension, treating all regions as functionally equivalent. We introduce the metabolism-weighted connectome (MwC), a framework that maps brain connectivity weighted by each region’s energy expenditure. Using simultaneous PET-MRI in healthy volunteers, we show that energetically most active hubs anchor networks for complex cognition, exhibit molecular signatures of intense synaptic activity, and are disproportionately vulnerable to Alzheimer’s disease. These findings establish a unifying principle: the brain regions most essential for cognition bear the greatest metabolic burden, and this cost may underlie their susceptibility to neurodegeneration.

Functional connectivity (FC) from resting-state fMRI captures temporal correlations between brain regions but cannot reveal the direction of neural signalling. Determining effective connectivity, the influence of one neural system over another, is essential for understanding cortical hierarchy and its energetic constraints. We extend Metabolic Connectivity Mapping (MCM; Riedl et al., 2016), a biologically grounded framework that infers directionality by integrating FC with glucose metabolism measured via [ ¹⁸ F]fluorodeoxyglucose positron emission tomography ([ ¹⁸ F]FDG PET). MCM builds on the principle that postsynaptic neurons consume more energy than presynaptic ones (Attwell and Laughlin, 2001; Attwell and Gibb, 2005), linking higher local metabolism to afferent input. Here, we present a new whole-cortex implementation that estimates directed connectivity directly from inter-regional energy ratios, enabling application to multimodal and fMRI-only datasets using an average cerebral metabolic rate of glucose (CMR _Glc ) map. The model reproduces hierarchical signalling within visual and sensorimotor systems and identifies novel directional asymmetries along sensory-cognitive gradients. MCM-derived metrics correlate with independent biological markers, including mitochondrial density (Mosharov et al., 2025) and cortical cytoarchitecture indexed by cell layer profiles (Amunts and Zilles, 2015; Wagstyl et al., 2020). By decomposing functional connectivity into metabolically constrained directed and undirected components, this framework bridges the gap between statistical connectivity and neuroenergetic mechanisms. Our results position MCM as a scalable and biologically interpretable model for inferring directed brain connectivity from human neuroimaging data.