Mapping the neural basis of selected cognitive functions: A combined functional, structural, and diffusion MRI study

Complex neuronal network interactions underlie cognitive processes, enabling the brain to adapt effectively to the external environment. Advanced neuroimaging techniques have facilitated the identification of potential targets and relevant endophenotypes for diagnosis and rehabilitation purposes. This study aims to explore the neuroanatomical correlation of various cognitive tasks using a combination of functional, structural, and diffusion MRI data to to characterize how brain regions across multiple modalities covary with cognitive performance.

Three hundred healthy adults from the IBID cohort completed a 15-test neuropsychological battery spanning memory, visuospatial ability, executive control, decision-making and processing speed. Structural MRI, diffusion MRI and resting-state fMRI were processed to derive gray-matter VBM maps, fractional anisotropy and intrinsic connectivity in MNI space; voxelwise regressions with cognitive scores were followed by total/combined maps and multimodal fusion using non-parametric combination and joint ICA, with atlas-based, FDR-corrected ROI correlations quantifying and localizing multimodal brain-cognition associations.

Single-modality analyses of gray matter, white matter and resting-state fMRI showed the largest voxel involvement in the left thalamus, left cerebellum, left superior temporal gyrus, right middle frontal gyrus and bilateral cingulate cortex. Multimodal fusion and FDR-corrected ROI analyses further indicated that middle frontal gyri, cingulate cortex, insula and superior/inferior parietal lobules were most strongly related to executive and speeded tasks (TMT-A/B, Stroop, SDMT, N-back, verbal fluency), whereas hippocampus, parahippocampal gyrus, posterior cingulate cortex and precuneus were selectively associated with episodic memory performance (RAVLT, Benson).

Taken together, these findings suggest that integrating structural, diffusion, and resting-state fMRI provides a nuanced but strictly descriptive view of how gray-matter morphology, white-matter microstructure, and intrinsic functional connectivity covary with performance across multiple cognitive domains in healthy adults. The resulting multimodal patterns are best regarded as a normative scaffold for future longitudinal and clinical studies of brain-cognition coupling, rather than as direct evidence for diagnostic utility or specific therapeutic interventions.