Neural networks form the basis for communication and information processing in the central nervous system. The human brain is capable of massive serial and parallel processing of big data enabling complex behavior and interaction with a constantly changing and challenging environment. To this end, neurons are organized on micro, meso and macroscale levels. The hierarchical and modular organization of neurons into ensembles enables segregation, and subspecialization, and also the integration of function across different temporospatial scales. The underlying morphology-activity, i.e. structure-function, relationships are embedded within the brain connectome. The connectome constitutes the substrate for self-organizing, emergent complex behavior that can dynamically adapt to a changing environment through synaptic plasticity, necessary for memory and learning. Neural network communication is governed by Hebbian rules, while maintenance of normal function involves homeostatic plasticity rules. Any perturbation is likely to trigger plastic responses at different scales governed by an interplay of both homeostatic and Hebbian plasticity.
A highly interesting perspective is the ability to understand emerging neural network behavior on the basis of the spontaneous dynamic state of self-organized criticality (SoC). SoC is a universal attribute of neural systems and is characterized by maximum information transmission and computational capacity. Once a network has reached the SoC state, neuronal avalanches can be observed, which represent cascades of activity whose size distribution is determined by power law. Neuronal avalanches can be observed in neural networks in vitro as well as in vivo however the exact mechanisms regulating such phenomena, not least their relevance to brain oscillations or the role of such critical states in the context of normal neural network function as well as adaptive or maladaptive responses to perturbations remain unclear. Of particular interest is whether such dynamics may be driven by specific network topologies, and to what extent they may be interpreted to infer the computational dynamics of the network within the different temporospatial scales of the brain connectome. In this context, the study and elucidation of related phenomena, including phase transitions, (stochastic) oscillations, synaptic transmission and small world organization of neural networks are of fundamental importance. Equally intriguing is whether neural network criticality can be harnessed, not least in the absence of directly observable behavioral correlates, to distinguish between adaptive and maladaptive responses to perturbations and harness or induce experience-dependent plasticity in a manner that can restore lost function.
The aim of this Research Topic is to summarize the current state-of-the-art in the context of key conceptual, methodological, and analytical tools applied to study and understand criticality in neural networks within the context of the connectome. This includes emergent behavior such as memory and cognition, dynamic morphology-activity relationships at the micro, meso, and macroscale in response to perturbations, as for example, trauma or neurodegenerative disease, as well as experienced-dependent plasticity and gain of function in response to rehabilitation.
Topics of interest include, but are not limited to:
• In vitro modeling of self-organizing, emerging behavior in neural networks using MEAs; neuromorphic systems; brain slice electrophysiology
• Electrophysiology during navigational task execution comparing normal versus lesioned animals; combination with chemo- and optogenetic tools
• Computational modeling of in vitro neural networks; state inference, prediction of phase transitions of emerging behavior in unperturbed and perturbed conditions
• Connectome function and structure; adaptive and maladaptive plasticity in health and disease (e.g., spinal cord injury, stroke, Alzheimer´s disease, ALS, epilepsy)
• Computational modeling; graph theory; recurrent neural networks
Neural networks form the basis for communication and information processing in the central nervous system. The human brain is capable of massive serial and parallel processing of big data enabling complex behavior and interaction with a constantly changing and challenging environment. To this end, neurons are organized on micro, meso and macroscale levels. The hierarchical and modular organization of neurons into ensembles enables segregation, and subspecialization, and also the integration of function across different temporospatial scales. The underlying morphology-activity, i.e. structure-function, relationships are embedded within the brain connectome. The connectome constitutes the substrate for self-organizing, emergent complex behavior that can dynamically adapt to a changing environment through synaptic plasticity, necessary for memory and learning. Neural network communication is governed by Hebbian rules, while maintenance of normal function involves homeostatic plasticity rules. Any perturbation is likely to trigger plastic responses at different scales governed by an interplay of both homeostatic and Hebbian plasticity.
A highly interesting perspective is the ability to understand emerging neural network behavior on the basis of the spontaneous dynamic state of self-organized criticality (SoC). SoC is a universal attribute of neural systems and is characterized by maximum information transmission and computational capacity. Once a network has reached the SoC state, neuronal avalanches can be observed, which represent cascades of activity whose size distribution is determined by power law. Neuronal avalanches can be observed in neural networks in vitro as well as in vivo however the exact mechanisms regulating such phenomena, not least their relevance to brain oscillations or the role of such critical states in the context of normal neural network function as well as adaptive or maladaptive responses to perturbations remain unclear. Of particular interest is whether such dynamics may be driven by specific network topologies, and to what extent they may be interpreted to infer the computational dynamics of the network within the different temporospatial scales of the brain connectome. In this context, the study and elucidation of related phenomena, including phase transitions, (stochastic) oscillations, synaptic transmission and small world organization of neural networks are of fundamental importance. Equally intriguing is whether neural network criticality can be harnessed, not least in the absence of directly observable behavioral correlates, to distinguish between adaptive and maladaptive responses to perturbations and harness or induce experience-dependent plasticity in a manner that can restore lost function.
The aim of this Research Topic is to summarize the current state-of-the-art in the context of key conceptual, methodological, and analytical tools applied to study and understand criticality in neural networks within the context of the connectome. This includes emergent behavior such as memory and cognition, dynamic morphology-activity relationships at the micro, meso, and macroscale in response to perturbations, as for example, trauma or neurodegenerative disease, as well as experienced-dependent plasticity and gain of function in response to rehabilitation.
Topics of interest include, but are not limited to:
• In vitro modeling of self-organizing, emerging behavior in neural networks using MEAs; neuromorphic systems; brain slice electrophysiology
• Electrophysiology during navigational task execution comparing normal versus lesioned animals; combination with chemo- and optogenetic tools
• Computational modeling of in vitro neural networks; state inference, prediction of phase transitions of emerging behavior in unperturbed and perturbed conditions
• Connectome function and structure; adaptive and maladaptive plasticity in health and disease (e.g., spinal cord injury, stroke, Alzheimer´s disease, ALS, epilepsy)
• Computational modeling; graph theory; recurrent neural networks