Does the cortex truly operate at criticality?*

It has been hypothesized that living networks of cortical neurons operate at a critical point, exactly between a phase where cascading activity would be damped and a phase where it would be amplified. Many experiments have supported this hypothesis, and have shown peaks in information transmission, dynamic range and susceptibility near a critical point. Despite this evidence, other experiments seem to suggest that the cortex operates in a slightly sub-critical regime. Thus, controversy remains over whether the cortex operates at criticality. In this talk, I will describe how spontaneous neuronal activity will prevent true criticality from being attained in these networks. I will show how this "quasi-critical" state differs from true criticality, and yet still allows near-optimal information processing. I also will describe testable predictions of quasi-criticality that could be evaluated soon. (Joint work done with Rashid Williams-Garcia and Gerardo Ortiz).