Skip to content

Criticality Creates a Functional Platform for Network Transitions Between Internal and External Processing Modes in the Human Brain

Minkyung Kim, Hyoungkyu Kim, Zirui Huang, George A. Mashour, Denis Jordan, Rüdiger Ilg, UnCheol Lee

Frontiers in Systems Neuroscience December 1, 2021 DOI: 10.3389/fnsys.2021.657809 via DOAJ

Summary

AI-generated from the abstract

The brain's ability to switch between internal and external modes is crucial for generating models of self and world, and this switching may rely on a state near criticality—a balanced condition between order and disorder. Large synchronization fluctuations of brain networks near criticality create temporal windows that favor either integrating internal information or processing external stimuli. Using computational modeling, EEG, and fMRI analyses across altered states of consciousness, synchronized networks bias toward internal information while incoherent networks bias toward external information. These preferences are most prominent at criticality and in conscious states associated with 4–12 Hz bandwidth.

Study at a glance

Characteristics Computational modeling and observational study with EEG and fMRI analysis Peer reviewed
Population Humans in various states of consciousness (general anesthesia, psychedelic states, minimally conscious states, unresponsive wakefulness syndrome)
Keywords Criticality Consciousness Oscillator model EEG FMRI
Key finding Near criticality, synchronized brain networks favor internal information integration and incoherent networks favor external information processing, with these preferences most prominent in conscious states and disrupted in unconscious states.

Abstract

Continuous switching between internal and external modes in the brain appears important for generating models of the self and the world. However, how the brain transitions between these two modes remains unknown. We propose that a large synchronization fluctuation of brain networks, emerging only near criticality (i.e., a balanced state between order and disorder), spontaneously creates temporal windows with distinct preferences for integrating the network’s internal information or for processing external stimuli. Using a computational model, electroencephalography (EEG) analysis, and functional magnetic resonance imaging (fMRI) analysis during alterations of consciousness in humans, we report that synchronized and incoherent networks, respectively, bias toward internal and external information with specific network configurations. In the brain network model and EEG-based network, the network preferences are the most prominent at criticality and in conscious states associated with the bandwidth 4−12 Hz, with alternating functional network configurations. However, these network configurations are selectively disrupted in different states of consciousness such as general anesthesia, psychedelic states, minimally conscious states, and unresponsive wakefulness syndrome. The network preference for internal information integration is only significant in conscious states and psychedelic states, but not in other unconscious states, suggesting the importance of internal information integration in maintaining consciousness. The fMRI co-activation pattern analysis shows that functional networks that are sensitive to external stimuli–such as default mode, dorsal attentional, and frontoparietal networks–are activated in incoherent states, while insensitive networks, such as global activation and deactivation networks, are dominated in highly synchronized states. We suggest that criticality produces a functional platform for the brain’s capability for continuous switching between two modes, which is crucial for the emergence of consciousness.

Comments

No comments yet.

Log in to comment