Long-term stability of avalanche scaling and integrative network organization in prefrontal and premotor cortex

Ongoing neuronal activity in the cortex establishes functional networks of synchronization that reflect normal and pathological brain function. The reconstruction of these networks typically suffers from the use of indirect measures of neuronal activity at low spatiotemporal resolution and a lack of longitudinal tracking. Accordingly, the precise nature of the underlying synchronization dynamics and its translation into robust graph theoretical markers are not well characterized. Here, we studied the stability of cortical dynamics and reconstructed functional networks over many weeks in prefrontal and premotor cortex of awake nonhuman primates. We monitored neuronal population activity directly in the ongoing local field potential (LFP) at high spatial and temporal resolution using chronically implanted high-density microelectrode arrays. Ongoing activity was composed of neuronal avalanches exhibiting stable, inverted parabolic profiles with the collapse exponent of 2 in line with a critical branching process. Avalanche-based functional networks, reconstructed using a Normalized Count estimator, revealed robust integrative properties characterized by high neighborhood overlap between strongly connected nodes and robustness to weak-link pruning. "Entropy of mixing" analysis demonstrated progressive link reorganization over weeks. The long-term stability of avalanche scaling and integrative network organization should support the development of robust biomarkers to characterize normal and abnormal brain function.