Neural Network Decorrelation for Healthy Brain Aging: A Cross-sectional and Longitudinal

Magnetoencephalography (MEG)

A noninvasive technique that detects magnetic fields above the surface of the head produced by postsynaptic potentials in the brain.

Society for Neuroscience - 2016-01-01James L, Leuthold A, Georgopoulos A, Chorn C, Mathison JH, Georgopoulos AP
Neural network decorrelation is fundamental to information processing. Specifically, ensemble freedom is constrained by correlations amount network elements: a network with least correlated elements provides maximum independence (i.e., zero mutual information) and hence entails maximum possibilities for encoding information.We have previously demonstrated that neural decorrelation, primarily involving temporal regions, distinguishes healthy veterans from those with psychiatric disorders, and have hypothesized that network decorrelation underlies healthy brain functioning by permitting neural flexibility (James et al., JAMA Psychiatry 70:410-418). In the present study (, we tested the hypothesis that in cognitively healthy individuals, decorrelation would increase with age, serving as a mechanism that promotes efficient information processing and neural flexibility to maintain healthy brain functioning across the lifespan.