A central goal of neuroscience is to understand how the brain synthesises information into a unified conscious experience. Here, we address two fundamental questions: how is the human information-processing architecture organised, and how does it support consciousness? Developing an information-resolved approach to functional connectivity in the human brain, we reveal that sensorimotor processing is supported by redundant interactions, whereas integrative processes rely on synergistic information, which is more prevalent in humans than non-human primates, with high- synergy regions exhibiting the highest degree of evolutionary cortical expansion and synaptic density. We delineate a "synergistic global workspace" architecture comprising gateway regions that gather information from specialised modules, which is then integrated within the workspace and widely distributed via broadcaster regions. Through functional MRI analysis, we show that gateway regions of the workspace correspond to the brain's default network, whereas broadcasters coincide with the executive control network.Remarkably, loss of consciousness due to anaesthesia or brain injury corresponds to reduced integrated information between gateway regions of the synergistic workspace, which is restored upon recovery. Thus, loss of consciousness may coincide with a breakdown of information integration accessing the synergistic workspace of the human brain, providing an avenue to reconcile prominent theoretical accounts of consciousness.
Keywords: #Consciousness #Mathematical_Models_Of_Consciousness #Luppi #Neuroscience