Fig. 2: Association of memory capacity with measures of structural integrity, functional activation and locus coeruleus intensity.
From: Computational memory capacity predicts aging and cognitive decline

A Spatial maps of the ICA components corresponding to (from top to bottom) medial visual, lateral visual, sensorimotor, temporal, right attention, and dorsal attention networks (left). The colored areas in the right panel showed significant positive association between global memory capacity and functional connectivity within the corresponding networks. Only networks that showed significant associations after correcting for multiple comparisons are shown overlaid on the MNI152 structural template93,94,95. B Visualization of white matter tracts that showed positive association between global memory capacity and FA values. The underlying FA template was created from the average of MNI-registered FA volumes from 20 representative Cam-CAN subjects under 36 years of age. We also used the tbss_fill command in FSL on the FA results, which thickens the thresholded stats image and allows for better visualization of the results. C Regions that had significant interaction between regional memory capacity and age in a linear model where the intensity of the locus coeruleus was used as a dependent variable. The average gray matter probability map (MNI152 template)93,94,95,96 shows only the regions with significant interactions after correcting for multiple comparisons (FDR < 0.05) in red. Plot of locus coeruleus intensity as a function of age for superior frontal gyrus, for different levels of memory capacity measured across the sample (level 1: mean memory capacity, yellow; level 2: 2 standard deviations above the mean, gray; level 3: 2 standard deviations below the mean, red). The plotted values are the z-scores of the corresponding raw locus coeruleus intensity values, shaded areas indicate 95% CI for predictions. Source data for Fig. 2 is provided as a Source Data file.