Brain-state dynamics in healthy aging

Aging is associated with alterations in brain structure and function. The reorganisation of the functional networks in aging should manifest as changes in the ongoing brain dynamics, even without changes in behaviour. Functional brain networks reorganise, either by slow changes in the relative contribution or rewiring of brain areas (plasticity) or via fast modulation of their causal interactions (effective connectivity). This reorganisation compensates for the structural or functional changes associated with aging and neurodegeneration. There is a growing interest in analysing the brain’s dynamical regimes, across multiple temporal length-scales, over which the brain networks are assumed to switch, from few milli-seconds to seconds. The aim to identify and characterise transiently stable and recurrent patterns of activity in the ongoing M/EEG (magneto/electroencephalography) at topography, sources, or functional connectivity levels. We characterized changes in temporal dynamics of brain states due to aging. Using Hidden semi-Markov models, with a log-normal duration distribution, we identified 7 brain states and explored group differences between younger and older adults. Five minutes of ongoing EEG-RS data from older adults > 60 (n=26) and young adults (18- 40) (n=26) was used to train the HsMM model. All the participants were right-handed and without cognitive impairment. We used the trained HSMM to extract individual state fractional occupancy (FO), state mean duration, and duration distribution tail and compared between groups. Mean duration and variance were significantly different in 6 out of the 7 brain states, while FO was different in only one state. Mean duration and duration distribution tail were longer in young adults than in older adults. Brain dynamics provides rich information that could help understanding the structural and functional changes associated with aging, even when no behavioural alterations are observed. More work is needed to understand the neurophysiological correlates of Brain States inferred from the modelling tools.