How does a low-level sensory representation rise into conscious awareness? What determines whether higher-order cortical areas have access to the representational content in sensory regions? Although much evidence suggests prefrontal cortical regions are critically responsible for awareness, assessing trial-by-trial connectivity from prefrontal to lower-level sensory areas via fMRI is challenging with conventional approaches. In this talk I will present a new dynamic functional connectivity metric, based on information theory, that we are developing. The metric relies on 40-way multinomial sparse logistic regression decoding, capitalizing on a 38-subject database for which voxel-based representational patterns have been functionally aligned via Procrustean transformations. This metric may be used to assess trial-by-trial functional connectivity between prefrontal and low-level sensory areas in real-time fMRI contexts, thus providing a tool for rigorously testing the role of prefrontal cortex in conscious awareness.