The Critical Brain Hypothesis in Different Physiological Settings
Date
2024-01
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Observations of neural activity across scales and species have often been found to display diverging correlations and scale-free collective dynamics in the form of information cascades, also called ‘neuronal avalanches’. These findings have motivated the critical brain hypothesis which posits that brain dynamics are self-tuned (e.g., perhaps via excitation-inhibition balance) to a second-order phase transition (or critical point), that separates exponentially-growing dynamics from quiescent states, to achieve optimality. At criticality, the
physics of phase transitions predict various scale-free quantities and macroscopic observables that depend only on a few properties, such as network symmetries and neurons dynamics, but not on the exact connectivity between neurons. The ubiquity of signatures of criticality suggest this critical state could be related to ‘normal’ brain function. However, this relationship is not yet fully understood, and most of what is known is only in the context of spontaneous or resting-state brain dynamics. This thesis explores how different physiological states of the subject affect signatures of criticality. It is shown that measures of criticality are susceptible to the natural heterogeneity between neurons in the response to a stimulus. This can skew observed statistics and could potentially explain current difficulties in establishing criticality in behaviour. It is also found that signatures of cortical desynchronization associated with the rapid eye movement state of sleep, spread across the cortical surface in a scale-free manner. This feature is absent if spatial information is excluded, emphasizing the importance of considering spatial connectivity when addressing questions of criticality. Finally, it is shown that while low-dose anesthetics do not significantly alter the critical state, surgical-plane anesthesia is non-universal, reflecting the mechanisms of said anesthetic. This fills a crucial knowledge gap as it was not know prior to this if deviations from criticality observed across previous experiments indeed reflected the anesthetic or trivially reflected differences in experimental setups. These results show how physiological state changes may result in deviations away from criticality, and further our understanding of the relationship between criticality and normal brain function. It is also found that signatures of cortical desynchronization associated with the rapid eye movement state of sleep, spread across the cortical surface in a scale-free manner. This feature is absent if spatial information is excluded, emphasizing the importance of considering spatial connectivity when addressing questions of criticality. Finally, it is shown that while low-dose anesthetics do not significantly alter the critical state, surgical-plane anesthesia is non-universal, reflecting the mechanisms of said anesthetic. This fills a crucial knowledge gap as it was not know prior to this if deviations from criticality observed across previous experiments indeed reflected the anesthetic or trivially reflected differences in experimental setups. These results show how physiological state changes may result in deviations away from criticality, and further our understanding of the relationship between criticality and normal brain function.
Description
Keywords
Criticality, Phase Transitions, Brain, Dynamics
Citation
Curic, D. (2024). The critical brain hypothesis in different physiological settings (Doctoral thesis, University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca.