Lars Penke, Susana M Maniega, Catherine Murray, Alan J Gow, Maria C Valdes Hernandez, Jonathan D Clayden, John M Starr, Joanna M Wardlaw, Mark E Bastin, and Ian J Deary (2010)

A general factor of brain white matter integrity predicts information processing speed in healthy older people

Journal of Neuroscience, 30:7569-7574.

Human white matter integrity has been related to information processing speed, but it is unknown whether impaired integrity results from localized processes or is a general property shared across white matter tracts. Based on diffusion MRI scans of 132 healthy individuals with a narrow age range around 72 years, the integrity of eight major white matter tracts was quantified using probabilistic neighborhood tractography. Principal component analyses (PCAs) were conducted on the correlations between the eight tracts, separately for four tract-averaged integrity parameters: fractional anisotropy, mean diffusivity, and radial and axial diffusivity. For all four parameters, the PCAs revealed a single general factor explaining similar to 45\% of the individual differences across all eight tracts. Individuals' scores on a general factor that captures the common variance in white matter integrity had significant associations with a general factor of information processing speed for fractional anisotropy (r = -0.24, p = 0.007) and radial diffusivity (r = 0.21, p = 0.016), but not with general intelligence or memory factors. Individual tracts showed no associations beyond what the common integrity factor explained. Just as different types of cognitive ability tests share much of their variance, these novel findings show that a substantial amount of variance in white matter integrity is shared between different tracts. Therefore, impaired cortical connection is substantially a global process affecting various major tracts simultaneously. Further studies should investigate whether these findings relate more to the role of tract integrity and information processing speed in nonpathological cognitive aging or in lifelong-stable processes.