| Session | ||
Symposium_06
| ||
| Presentations | ||
Brain Criticality and Consciousness Abstract A growing body of research suggests that the brain operates near criticality, the point at a phase transition between order and disorder. This concept is emerging as a promising framework for understanding diverse phenomena in neuroscience. This symposium will explore the role of measures of brain criticality to study consciousness. We will introduce the foundational concepts of criticality, detailing its theoretical underpinnings and relevance as a unifying framework in the neuroscience of consciousness. We will explore research showing how individual differences in levels of synchronisation which manifest through differences in cognition and consciousness can be mapped through brain states which occupy an extended critical regime yielding certain functional advantages. We will also present evidence from computational modelling across 103 mammal species showing that measures relating to brain criticality may convey information as to informational processing across species and its relationship to consciousness. We will finally discuss how deviations from a criticality may give clinically useful information as to the pathological/perturbed brain’s proximity to a conscious or healthy state, also by using measures of proximity to a first order phase transition we can detect the speed of recovery of consciousness from anaesthesia. Bringing together experts from diverse fields, the session aims to shed light on how criticality — the state of being poised between order and disorder — might provide insights into the mechanisms underlying consciousness. Rationale on symposium's general scientific interest By bringing ideas from physics and complexity science to diverse methodologies, including electrophysiology, neuroimaging, computational modelling, and neuromodulation, this symposium addresses at several domains that are relevant to attendees of the ASSC. We bring together researchers from a diverse set of fields, offering their perspective from not only within the heart of the science of consciousness, but also introducing ideas about how criticality underwrites efficient information processing in cognition and more broadly across species, thus creating a cohesive and comprehensive view of how this framework can uniquely enhance our understanding of brain function. Rationale on complementarity of talks Each talk of the symposium gives a unique perspective on brain criticality providing a progression from theoretical foundations to electrophysiology and neuromodulatory work to functional magnetic resonance imaging in humans and computational modelling across species. While the symposium showcases brain criticality, we also aim to captivate a broad audience interested in consciousness science. The presentations provide a comprehensive view of the criticality framework’s applications, from neurological conditions like disorders of consciousness and epilepsy to pharmacological perturbations such as psychedelics and anaesthesia, alongside broader neuroscientific considerations for optimal information processing supporting cognition. Rationale on timeliness/importance Experimental support for the brain criticality hypothesis began two decades ago, gaining recent momentum due to large-scale data and interdisciplinary methods from physics and complexity science. Evidence now highlights criticality’s potential as a unifying framework in neuroscience, appearing frequently in publications across various fields. Criticality is also becoming relevant for clinical efforts to characterise conditions like disorders of consciousness, epilepsy, and schizophrenia. This symposium aims to provide a comprehensive overview of brain criticality, and aims to appeal to both those with a passing interest in the increasingly popular topic and those who wish to delve deeper into the methodological formulations. Rationale on panel inclusivity We are researchers from multiple career stages, including two PhD students: Naji Alnagger, a UK born, middle-eastern researcher based in Belgium who organised the symposium and who will act as the chair, and Canadian-based researcher Jordan O’Byrne who will provide the vital opening presentation of the topic. We also have Gustavo Deco who for the past 20 years has contributed some of the most important advances in computational neuroscience. Satu Palva, is a senior researcher at the University of Glasgow, she has been a principal investigator since 2011. UnCheol Lee is a Korean associate professor currently at the University of Michigan. Presentations of the Symposium Phase Transitions and the Emergence of Typical and Atypical Consciousness Consciousness is thought to emerge from the collective action of billions of interacting neurons. Statistical physics offers a framework for bridging this gap between the microscopic multitude and the macroscopic whole. By virtue of their universality property, simple statistical physical models that were first developed to describe the emergent behaviour of condensed matter also happen to describe the behaviour of much more complicated systems, including plate tectonics, flocking birds and – as is becoming increasingly clear – the brain. Central to these models is the concept of a phase transition, whereby changes in the system’s macroscopic behaviour are brought about by tuning a single microscopic parameter. At the cusp of these transitions, systems are said to be at criticality, and take on an array of adaptive computational features such as input sensitivity and pattern diversity, features which would appear favourable to waking consciousness. In this talk, I will briefly review the body of work suggesting that brain criticality is a prerequisite of waking consciousness, including our recent TMS-EEG study tying brain perturbational complexity to criticality. I will then show new magnetoencephalography-based evidence suggesting that certain atypical states of consciousness, either pathological (schizophrenia) or pharmacologically-induced (the psychedelic state), can be explained by shifts of the brain’s dynamical state point with respect to different critical phase transitions and concomitant smoothing or roughening of the brain’s energy landscape. I aim to show that statistical physics offers an intuitive, mechanistic and conceptually transparent framework for linking alterations of brain activity with alterations of consciousness. Critical Dynamics of Network Oscillations Network-oscillations are fundamental for human cognition and consciousness. Especially, phase synchronization of neuronal oscillations has been thought to be a mechanism for routing communication across neuronal circuits in support of cognitive subfunctions. Yet, there is wide variability of both individual levels of coupling as well as their spatio-temporal patterns influencing behavior and cognitive capacity of which underlying mechanisms has remained not understood. We have recently shown that the variability in inter-individual levels of synchronization can be explained in the framework of brain criticality by individual brain states along an extended critical-like dynamics – the Griffiths Phase (Fusca et al., 2020). The position i.e. the operating point also varied between frequencies and brain areas. I will discuss these results and propose that the extended critical regime yields functional advantages of criticality over a wider range of values leading to individual, circuit and cognitive differences. Fuscà M, Siebenhühner F, Wang SH, Myrov V, Arnulfo G, Nobili L, Palva JM, Palva S (2023) Brain criticality predicts individual levels of inter-areal synchronization in human electrophysiological data. Nat Commun. 2023;14(1):4736. doi: 10.1038/s41467-023-40056- The Critical Behavior of the Mammalian Brain: Inferring Functional Cognitive Capabilities across Species from Anatomy We studied the functional predictions of brain connectivity matrices (connectomes), estimated using diffusion MRI, from 103 mammal species whose brain sizes vary over more than four orders of magnitude. For this, we built a brain model to generate collective activity from each connectome. The model was composed of binary units interacting through the connectome and presented a critical point associated to a phase transition between order and disorder. It is known that, at the critical point, brain computations are optimized (in terms of information transmission, storage, and processing) and several previous studies suggest that the spontaneous brain activity operates at this point’s vicinity. Following this view, we studied the critical values of two quantities that describe the functional capabilities of the network, namely the diversity of the repertoire of network states and the susceptibility of the network to an external stimulus. We found that both measures positively correlate with the logarithm of the brain’s volume and with the phylogenetic evolution. These results suggest that connectomes with larger sizes and phylogenetically more evolved, produce a larger repertoire of network states together with larger responses to external stimuli. Brain Networks’ Proximity to a First-Order Phase Transition Determines Early or Prolonged Recovery from Unconsciousness. Understanding why some patients recover quickly from pharmacologically or pathologically induced unconsciousness, while others take longer, is crucial for improving patient outcomes. Recent empirical and computational studies indicate that brain criticality is a necessary condition for the emergence of higher-order brain function, such as consciousness, and that measuring deviations from criticality provides insight into a perturbed brain's proximity to a conscious state. However, if brain criticality is a 'sweet spot' for the emergence of consciousness, it remains unclear why some brains lose and regain criticality more quickly or slowly under anesthesia or coma. We propose that a brain’s phase transition type—specifically, its proximity to a first-order phase transition—determines the recovery speeds of brain criticality and consciousness. Using computational modeling and empirical data from human and animal brains undergoing anesthesia- and coma model-induced state transitions, we demonstrate that brain networks closer to a first-order phase transition more easily lose criticality and recover more slowly due to higher instability at critical points. Furthermore, we show that the trajectories of consciousness loss and recovery can be systematically predicted and modulated by adjusting the brain network’s phase transition type between first- and second-order transitions. These findings have significant implications for designing resilient brain networks that can better withstand perturbations and recover quickly, with potential applications in facilitating conscious-like states in non-biological systems, such as artificial intelligence. | ||