Meta-analytic connectivity perturbation analysis (MACPA): a new method for enhanced precision in fMRI connectivity analysis

Co-activation of distinct brain areas provides a valuable measure of functional interaction, or connectivity, between them. One well-validated way to investigate the co-activation patterns of a precise area is meta-analytic connectivity modeling (MACM), which performs a seed-based meta-analysis on task-based functional magnetic resonance imaging (task-fMRI) data. While MACM stands as a powerful automated tool for constructing robust models of whole-brain human functional connectivity, its inherent limitation lies in its inability to capture the distinct interrelationships among multiple brain regions. Consequently, the connectivity patterns highlighted through MACM capture the direct relationship of the seed region with third brain regions, but also a (less informative) residual relationship between the third regions themselves. As a consequence of this, this technique does not allow to evaluate to what extent the observed connectivity pattern is really associated with the fact that the seed region is activated, or it just reflects spurious co-activations unrelated with it. In order to overcome this methodological gap, we introduce a meta-analytic Bayesian-based method, called meta-analytic connectivity perturbation analysis (MACPA), that allows to identify the unique contribution of a seed region in shaping whole-brain connectivity. We validate our method by analyzing one of the most complex and dynamic structures of the human brain, the amygdala, indicating that MACPA may be especially useful for delineating region-wise co-activation networks.