Lifespan driven reorganization of the global network dynamics unfold on a multifrequency landscape
Cognitive processes are mediated by communication among multiple brain areas that individually exhibit oscillatory neuro-electromagnetic signals emerging from a cascade of complex physiological processes. Several studies have reported that aging changes the intrinsic properties of neural oscillations, both in resting state and in the context of cognitive tasks. For example, the amplitude of resting and motor-related beta band oscillations (16-25 Hz) is typically found to be higher in the older
... igher in the older population compared to the younger population. Similarly, a substantial number of reports have highlighted that peak alpha frequency (8-12 Hz) is lowered in neurodegenerative disorders and in healthy aging. How such observations emerge from underlying neuronal signal processing mechanisms remains elusive. Furthermore, how do the spatiotemporal organization of these rhythms support the current neurobiological theories of aging is poorly understood. Here we addressed these issues using the resting state magnetoencephalogram (MEG) data from a large cross-sectional cohort consisting of 650 human participants with age range 18-90 covering the entire adult lifespan. Concurring with previous research in smaller cohorts, we found a consistent increase in the power of beta oscillations and a decrease in the peak alpha frequency as a function of age. Subsequently, we found significant posterior to anterior shifts in the spectral topographies of both alpha and beta bands. To reconcile these observations with a comprehensive theory at the level of network level signal processing, we computed the whole-brain global coherence that captures the degree of communication among nodes of a large-scale network as a function of aging. The global coherence among MEG signals increased with age in slower time-scales i.e. delta (1-3 Hz) and theta (3-7 Hz) frequencies. Simultaneously, global coherence decreased for faster timescales i.e. alpha (8-12 Hz) and beta (16-25 Hz) frequencies. Further, using the measure of metastability that quantifies the divergence of a network from a stable synchronous state, we characterized the dispersion of information processing in different frequency bands. Putting together, our study reveals how neurobiological theories of aging such as posterior to anterior shifts of sensory and cognitive processing, dynamic workspace hypothesis can all be reconciled using resting-state MEG data. We could highlight how the temporal structure of MEG signals is representative of a more comprehensive understanding of large-scale network mechanisms that govern lifespan dynamics.