Whole-brain causal connectivity during decoded neurofeedback: a meta study

Decoded Neurofeedback (DecNef) represents a pioneering approach in human neuroscience that enables modulation of brain activity patterns without subjective conscious awareness through the combination of real-time fMRI with multivariate pattern analysis. While this technique holds significant potential for clinical and cognitive applications, the causal mechanisms underlying successful DecNef regulation and the neural dynamics that distinguish successful learners from those who struggle remain poorly understood. To address this question, we conducted a meta-study across functional magnetic resonance imaging (fMRI) data from five DecNef experiments, each with multiple fMRI sessions, to reveal causal network dynamics associated with individual differences in neurofeedback performance. Using the newly proposed CaLLTiF causal discovery method, we computed causal maps to identify causal network patterns that distinguish DecNef regulation from baseline and account for variations in neurofeedback success. We found that enhanced connectivity within the bilateral control network--particularly stronger connections involving the posterior cingulate and precuneus cortex--predicted neurofeedback success across all five studies. Whole-brain causal connectivity during DecNef further exhibited distinct network reorganizations, characterized by reduced average path lengths and increased right-limbic nodal degrees. Further, comparisons across cognition- and perception-targeted DecNef revealed a remarkable separation in connections to and from the somatomotor network, where connections between somatomotor and control-default-attention networks are larger during cognitive neurofeedback while causal effects between somatomotor and subcortical-visual-limbic networks are larger during perceptive DecNef. This is despite the fact that none of the involved studies targeted or involved motor activity. Overall, our results demonstrated the key role of bilateral medial control network in successful DecNef regulation regardless of the DecNef targets, a clear separation in somatomotor involvement between cognitive and perceptive DecNef, and general promise of whole-brain causal discovery in understanding complex neural processes such as decoded neurofeedback.