Growing evidence shows that whole-brain functional connectivity (FC) temporarily settles into a finite repertoire of FC patterns during rest. However, the functional role of this spontaneous switching remains unclear and the best methods to characterize FC states are under debate.
I will present a novel approach to define FC states focusing on the dominant pattern of dynamic FC matrices captured by their leading eigenvector (Cabral et al., 2017). Leading eigenvector dynamics analysis (LEiDA) allows identifying the precise epochs when the FC becomes dominated by a different pattern, indicating a switch between FC states. Moreover, reducing the dimensionality of each NxN FC matrix to a vector with N elements improves the signal-to-noise ratio in the detection of recurrent patterns within and across sessions. Once recurrent FC patterns – or states – are detected with LEiDA, they can be characterized in terms of lifetime, probability of occurrence and switching profiles.
We used LEiDA to investigate how the switching behavior of resting-state FC relates with cognitive performance in healthy older adults. We analysed resting-state fMRI data from 98 healthy adults previously categorized as being among the best or among the worst performers in a cohort study of >1000 subjects aged 50+ who underwent neuropsychological assessment. We found that poorer cognitive performance is associated with weaker FC temporal similarity together with altered switching between FC states. These results provide new evidence linking the switching dynamics of FC during rest with cognitive performance in later life, reinforcing the functional role of resting-state activity for effective cognitive processing.
Cabral J, Vidaurre D, Marques P, Magalhaes R, Moreira PS, Soares JM, Deco G, Sousa N, Kringelbach ML (2017) Cognitive performance in healthy older adults relates to spontaneous switching between states of functional connectivity during rest. Scientific Reports. In Press