“The multi-timescale connectome”

NOTE corrected time

Hosted by the WashU Neuroimaging Community (WUNIC)

Abstract: 

The view of human brain function has drastically shifted over the last decade, owing to the observation that most brain activity is intrinsic rather than driven by external stimuli or cognitive demands. Specifically, all brain regions continuously communicate in spatiotemporally organized patterns that constitute the functional connectome, with consequences for cognition and behavior. In this talk, I will argue that another shift is underway, driven by new insights from synergistic interrogation of the functional connectome using different acquisition methods. The human functional connectome is typically investigated with functional magnetic resonance imaging (fMRI) that relies on the indirect hemodynamic signal, thereby emphasizing very slow connectivity across brain regions. Conversely, more recent methodological advances demonstrate that fast connectivity within the whole-brain connectome can be studied with real-time methods such as electroencephalography (EEG).

When combining fMRI with scalp or intracranial EEG in humans, especially when recorded simultaneously, we observed a “multiplex” of concurrent connectome processes. Specifically, the connectome concurrently comprises both fast, oscillation-based connectivity observable with EEG, as well as extremely slow processes best captured by fMRI. While the fast and slow processes comprise spatially similar patterns, they unfold in a temporally independent manner. The temporal dissociation suggests that infraslow and fast connectivity constitute multiple neurobiologically distinct processes. Each of these processes reflects a different dynamic trajectory through a shared state space of discrete connectome states. The multitude of flexible trajectories may enable any given brain region to concurrently connect to multiple sets of other regions, maximizing the brain’s capacity for information integration.

For inquiries contact Cathy Gezella.