On neural substrates of cognition: Theory, experiments and application in brain computer interfaces

Recent experiments with high-resolution brain imaging techniques provide an amazing view on the complex spatio-temporal dynamics of cortical processes. There is ample of evidence pointing to frequent transitions between periods of large-scale synchronization and intermittent desynchronization at alpha-theta rates (period length of 0.1 s to 0.25s). These observations have been interpreted based on the cinematic model of cognitive processing. In the corresponding mathematical theories, brains are perceived as open thermodynamic systems converting noisy sensory data into meaningful knowledge. We employ a graph-theoretic model called neuropercolation, which extends the concept of phase transitions to large interactive populations of nerve cells. We show that normal brains operate at the edge of criticality, where phase transitions are manifested via intermittent phase synchronization. Cortical phase transitions are viewed as neural correlates of cognition and serve as basis for non-invasive cognitive monitoring using novel brain-computer interfaces.