Abstract
The variation of brain functions as healthy ageing has been discussed widely using resting-state brain imaging. Previous conclusions may be misinterpreted without considering the effects of global signal (GS) on local brain activities. Up to now, the variation of GS with ageing has not been estimated. To fill this gap, we defined the GS as the mean signal of all voxels in the gray matter and systematically investigated correlations between age and indices of GS fluctuations. What’s more, these tests were replicated with data after hemodynamic response function (HRF) de-convolution and data without noise regression as well as head motion data to verify effects of non-neural information on age. The results indicated that GS fluctuations varied as ageing in three ways. First, GS fluctuations were reduced with age. Second, the GS power transferred from lower frequencies to higher frequencies with age. Third, the GS power was more evenly distributed across frequencies in ageing brain. These trends were partly influenced by HRF and physiological noise, indicating that the age effects of GS fluctuations are associated with a variety of physiological activities. These results may indicate the temporal dedifferentiation hypothesis of brain ageing from the global perspective.
Highlights
The variation of brain functions as healthy ageing has been discussed widely using resting-state brain imaging
Negative correlations between the SD of global signal (GS) and age were observed for original data (r = − 0.47, p < 0.001), but dramatically reduced for data without noise regression (r = − 0.14, p = 0.01), and disappeared for de-convolved data (r = − 0.06, p = 0.312), suggesting that the decline of GS variability with age may be caused by multiple factors including neural activity, physiological noise, and neurovascular coupling
The GS power was significantly correlated with age in multiple frequency bands [0.007–0.01 Hz (r = 0.12–0.24), 0.013–0.022 Hz (r = − 0.12 to − 0.20), 0.036–0.043 Hz (r = 0.12–0.13), 0.055–0.098 Hz (r = − 0.12 to − 0.26), and 0.118–0.25 Hz (r = 0.12–0.32); false discovery rate (FDR) corrected, q < 0.05 corresponding r = ± 0.12] for the original data (Fig. 1a, column 1 and 2)
Summary
The variation of brain functions as healthy ageing has been discussed widely using resting-state brain imaging. The GS power was more evenly distributed across frequencies in ageing brain These trends were partly influenced by HRF and physiological noise, indicating that the age effects of GS fluctuations are associated with a variety of physiological activities. These results may indicate the temporal dedifferentiation hypothesis of brain ageing from the global perspective. It has been demonstrated that the BOLD signal as well as other brain signals has two major characteristics: scale-free and oscillation[15,16] The former refers that the power of brain signals tends to fall off with increasing frequency following a power-law function, while the latter refers to the recurring pattern of brain activity that follows a particular temporal b eat[15]. Characteristics of GS with ageing which may contribute to the physiological mechanism of BOLD signal, hasn’t been studied yet
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
