Time varying connectivity in large scale networks in the human brain

It is well understood that the functioning of the human brain is based on a precise communication between a large number of brain regions. The functional communication also referred to as functional connectivity, is the result of coordination of spatially diverse and widely-distributed cortical brain regions. Moreover, the connectivity span over frequency and temporal dimensions and it is direction specific. Therefore, estimated functional connectivity should be directed, time varying and frequency dependent to fully uncover the functioning of the brain. Deriving time varying connectivity requires careful selection of the imaging modalities and the consideration of several technical steps. Compared to fMRI, EEG data have an excellent time resolution. Throughout this work, we discussed methods to calculate time varying connectivity from the sources of EEG data either measured during visual spatial attention tasks or generated with simulations. We specifically focused on the directed connectivity among areas. This allows understanding the influence of one brain area on the other areas and vice versa. We started with constructing a pipeline for source modelling from surface EEG data and for the calculation of time-varying connectivity from the estimated sources. This involved understanding and carefully selecting different source modelling algorithms and methods. We constructed a realistic head model using subject-specific anatomy and EEG channel locations. Standardized low-resolution brain electromagnetic tomography (sLORETA) was used to estimate the brain sources. For the connectivity analysis, we mainly focused on two approaches: (1) time-varying multivariate autoregressive modelling (TV-MVAR) of the data followed by calculation of partial directed coherence (PDC), a Granger Causality (GC) based multivariate spectral measure. (2) In a second approach, the phase lag index (PLI), a phase synchronisation based measure, was used. We also used PLI weighted with the cross-spectral power, referred to as weighted PLI (wPLI), to improve the noise performance. In the first study, we applied the weighted phase lag index based time varying connectivity framework to electrocorticography (ECoG) data acquired during a visual spatial attention task. We observed significant directed influences related to the invalidity and competition effects in the parietal cortex. We found that in invalid trials, the superior parietal lobule (SPL) was driving the activity in the intraparietal sulcus (IPS). Furthermore, when a competing distracter was present, a prolonged effect of anterior IPS to middle IPS regions was found. Overall, this suggests the origin of late selection in IPS at the postperceptual stage. The study provided for the first time the electrophysiological signature of spatial shifting in response to an invalid cued spatial trial in SPL. Based on time varying connectivity we were able to study the reversal of the directed influences something which is not possible with undirected time-invariant connectivity. Then, we moved a step further by comparing methods to estimate the time varying connectivity based on PDC. In these methods, the accurate estimation of the time varying MVAR model is an essential step. Kalman filtering based approaches such as the classical Kalman filter (CKF) and the General Linear Kalman filter (GLKF) offer a robust framework. However, there are a few ways in which Kalman filtering based techniques can be implemented when data are acquired in the form of multiple trials as is usually the case. We used simulated data with a predefined connectivity model and used a realistic approach to generate EEG data from EEG sources. We studied the impact of the number of trials and SNR on both approaches. Given the faster computation time, GLKF is the best choice in most cases except when the number of trials is too low. Unfortunately, a lower limit of trials cannot be set for GLKF as it depends on several other parameters like the number of channels. Furthermore, the extraction of time series in a region is an essential step before connectivity estimation. Due to the ill-posed nature of the source modelling problem, the sources are smooth. We showed that the dipole with time series showing the highest correlation with the region average time series also indicated the highest correlation with the ground truth time series and this dipole had the lowest localisation error compared to other strategies. Finally, we studied temporal networks related to a visual spatial attention experiment in the human brain. We used EEG data acquired in 14 healthy subjects. The role of the parietal lobe in visual spatial attention is well established. However, there exist anatomically and functionally dissociable regions in the parietal lobe. To uncover the mechanisms of attention, the underlying connectivity between parietal regions needs to be investigated. We applied EEG source modelling combined with time-varying connectivity estimation based on partial directed coherence. With this approach, we studied a directed network at high temporal resolution using pre-defined regions mainly in the parietal lobe.status: publishe