Directed functional connectivity of the default-mode-network of young and older healthy subjects

Abstract

Alterations in the default mode network (DMN) are associated with aging. We assessed age-dependent changes of DMN interactions and correlations with a battery of neuropsychological tests, to understand the differences of DMN directed connectivity between young and older subjects. Using a novel multivariate analysis method on resting-state functional MRI data from fifty young and thirty-one healthy older subjects, we calculated intra- and inter-DMN 4-nodes directed pathways. For the old subject group, we calculated the partial correlations of inter-DMN pathways with: psychomotor speed and working memory, executive function, language, long-term memory and visuospatial function. Pathways connecting the DMN with visual and limbic regions in older subjects engaged at BOLD low frequency and involved the dorsal posterior cingulate cortex (PCC), whereas in young subjects, they were at high frequency and involved the ventral PCC. Pathways combining the sensorimotor (SM) cortex and the DMN, were SM efferent in the young subjects and SM afferent in the older subjects. Most DMN efferent pathways correlated with reduced speed and working memory. We suggest that the reduced sensorimotor efferent and the increased need to control such activities, cause a higher dependency on external versus internal cues thus suggesting how physical activity might slow aging.

Velocities of Hippocampal Traveling Waves Proportional to Their Coherence Frequency

“…Our results demonstrated that the spikes fell outside the wavelet confidence region (defined in Torrence and Compo [18,19]), which were not included in the calculations. This confirmed that the findings were not influenced by wavelet-filtering artifacts [19,20].…”

“…For bivariate (pair coherence) calculations, we used the phase-lag-index (PLI) [20,21], and for multivariate coherence four-node pathways, we used the pathway-index (PWI) [10][11][12][13][14][15][16]22] (see method). Both calculations employed wavelet analysis with custom-developed software.…”

“…In our prior study, we showcased the utilization of interactions among four coupled time-series to determine their temporal coupling order at each time-frequency point [12,[15][16][17][18][19][20]. Two key assumptions underpinned these inferences: firstly, that the coupling between the time-series was relatively weak, and secondly, that propagation was predominantly unidirectional.…”

“…To this end, we conducted an analysis of traveling waves within local field potential (LFP) signals extracted from the human hippocampus in neurosurgical epileptic subjects undergoing stereoelectroencephalography (SEEG) electrode implantation. In contrast to previous studies employing various linear methods for measuring traveling waves, we utilized a novel multivariate nonlinear analysis method previously applied to functional MRI signals [12,[15][16][17][18][19][20]. This nonlinear analysis leverages mutual coherence (simultaneous coherence) among four time-series (i.e., LFP signals, each from a distinct location), along with the temporal delays at each frequency between these signals, to establish their temporal order and infer directed pathways.…”