Partial replication of stable individual factors in resting state functional brain networks

Functional brain organisation is widely studied today, however with diverging methods which lead to hard to compare results. For example, in modelling resting state functional networks, one may choose a hard-parcellation which assumes a fixed spatial map of brain areas which are temporally synchronous in their activity. On the other hand, soft-parcellations may assume a fixed temporal signal which can vary across spatial areas with different weights. In this modelling approach a single brain area may then belong to several networks. With such fundamental differences in models of functional organisation, questions may be raised as to whether certain results in the field are model dependent or translate across models. The following project is aimed at investigating stability of resting state functional brain networks at the individual level, partially replicating Gratton et al. (2018) though using a different dataset and different functional brain network model. Instead of a hard-parcellation of 9 densely sampled individuals (MSC; Gordon et al., 2017), we use the HCP test-retest dataset (n = 45) and estimate individual level networks with a soft-parcellation algorithm known as Probabilistic Functional Modes (PROFUMO; Harrison et al., 2015; 2020). Lastly, instead of comparing connectivity matrices of a network we compare the spatial topography of networks.