Aging in humans is associated with increased differences in cognitivefunction between individuals, which are largely attributed to variability in the brain.However, the fundamental concept regarding individual differencesin functional or anatomical characteristics in older individuals is not yet completely understood, and it isessential to evaluate the extent to which individual variability in the brain contributes to individual differences in cognitive ability in elderly individuals.Here we used MRI toinvestigate the inter-individual variability in gray matter density (GMD) and its correlation with cognitive aging.Eighty-eight healthy older volunteers (42 males and 46 females;age, 70.6±5.6 years) were included in the study.Voxel-based morphometry (VBM) was used to evaluate quantitatively the GMD differeces between individuals for each voxel.Individual cognitive performance was assessed regarding four functional domains, includingworking memory,episodic memory, executive functionand language using a battery of standardized neuropsychological tests.Results showed that individual GMD variability was heterogenousacross the brain, with significantly highest variability in the visual cortex and lateral parietal and lateral prefrontal cortices that constituted the executive-control and working memory network.However, significant correlations between GMD and cognitive performance was only found in a subcortical cluster that included the caudate, putamen,and the insula.The finding suggested that brain regions with largest individual GMD differences may not necessarily contributed greatest to individual differences in cognitive aging.It will be interesting in future studies to investigate whether the white matter fibres that connect neurons in differentcortical regions may function critically in cognitive performance in older age.