Artificial Astrocytes Improve Neural Network Performance

Physical Review E, 2019

We explore the consequences of introducing higher-order interactions in a geometric complex network of Morris-Lecar neurons. We focus on the regime where travelling synchronization waves are observed out of a first-neighbours based coupling, to evaluate the changes induced when higherorder dynamical interactions are included. We observe that the travelling wave phenomenon gets enhanced by these interactions, allowing the information to travel further in the system without generating pathological full synchronization states. This scheme could be a step towards a simple modelization of neuroglial networks.

Encyclopedia of Computational Neuroscience, 2020

Neural Plasticity, 2016

Glutamatergic gliotransmission, that is, the release of glutamate from perisynaptic astrocyte processes in an activity-dependent manner, has emerged as a potentially crucial signaling pathway for regulation of synaptic plasticity, yet its modes of expression and function in vivo remain unclear. Here, we focus on two experimentally well-identified gliotransmitter pathways, (i) modulations of synaptic release and (ii) postsynaptic slow inward currents mediated by glutamate released from astrocytes, and investigate their possible functional relevance on synaptic plasticity in a biophysical model of an astrocyte-regulated synapse. Our model predicts that both pathways could profoundly affect both short- and long-term plasticity. In particular, activity-dependent glutamate release from astrocytes could dramatically change spike-timing-dependent plasticity, turning potentiation into depression (and vice versa) for the same induction protocol.

Glia, 2019

Systems neuroscience is still mainly a neuronal field, despite the plethora of evidence supporting the fact that astrocytes modulate local neural circuits, networks, and complex behaviors. In this article, we sought to identify which types of studies are necessary to establish whether astrocytes, beyond their well-documented homeostatic and metabolic functions, perform computations implementing mathematical algorithms that sub-serve coding and higher-brain functions. First, we reviewed Systems-like studies that include astrocytes in order to identify computational operations that these cells may perform, using Ca 2+ transients as their encoding language. The analysis suggests that astrocytes may carry out canonical computations in a time scale of subseconds to seconds in sensory processing, neuromodulation, brain state, memory formation, fear, and complex homeostatic reflexes. Next, we propose a list of actions to gain insight into the outstanding question of which variables are encoded by such computations. The application of statistical analyses based on machine learning, such as dimensionality reduction and decoding in the context of complex behaviors, combined with connectomics of astrocyte-neuronal circuits, is, in our view, fundamental undertakings. We also discuss technical and analytical approaches to study neuronal and astrocytic populations simultaneously, and the inclusion of astrocytes in advanced modeling of neural circuits, as well as in theories currently under exploration such as predictive coding and energy-efficient coding. Clarifying the relationship between astrocytic Ca 2+ and brain coding may represent a leap forward toward novel approaches in the study of astrocytes in health and disease. Kira E. Poskanzer and Elena Galea contributed equally to this study.

Springer Series in Computational Neuroscience, 2019

There is growing excitement around glial cells, as compelling evidence point to new, previously unimaginable roles for these cells in information processing of the brain, with the potential to affect behavior and higher cognitive functions. Among their many possible functions, glial cells could be involved in practically every aspect of the brain physiology in health and disease. As a result, many investigators in the field welcome the notion of a Neuron-Glial paradigm of brain function, as opposed to Ramon y Cayal's more classical neuronal doctrine which identifies neurons as the prominent, if not the only, cells capable of a signaling role in the brain. The demonstration of a brain-wide Neuron-Glial paradigm however remains elusive and so does the notion of what neuron-glial interactions could be functionally relevant for the brain computational tasks. In this perspective, we present a selection of arguments inspired by available experimental and modeling studies with the aim to provide a biophysical and conceptual platform to computational neuroscience no longer as a mere prerogative of neuronal signaling but rather as the outcome of a complex interaction between neurons and glial cells.

2017 IEEE Computer Society Annual Symposium on VLSI (ISVLSI), 2017

This paper presents a hardware based implementation of a biologically-faithful astrocyte-based selfrepairing mechanism for Spiking Neural Networks. Spiking Astrocyte-neuron Networks (SANNs) are a new computing paradigm which capture the key mechanisms of how the human brain performs repairs. Using SANN in hardware affords the potential for realizing computing architecture that can self-repair. This paper demonstrates that Spiking Astrocyte Neural Network (SANN) in hardware have a resilience to significant levels of faults. The key novelty of the paper resides in implementing an SANN on FPGAs using fixed-point representation and demonstrating graceful performance degradation to different levels of injected faults via its self-repair capability. A fixed-point implementation of astrocyte, neurons and tripartite synapses are presented and compared against previous hardware floating-point and Matlab software implementations of SANN. All results are obtained from the SANN FPGA implementat...

Neuroscience & Biobehavioral Reviews, 2020

Astrocytes are heterogeneous population of neural cells with diverse structural, functional and molecular characteristics responsible for homeostasis and protection of the central nervous system (CNS). Unlike neurones, astrocytes do not generate action potentials, but employ fluctuations of cytosolic ions as a substrate for their excitability. Ionic signals are associated with neuronal activity and these signals initiate an array of responses ranging from the activation of plasmalemmal homeostatic transporters to the secretion of numerous signalling molecules including neuromodulators, neurotransmitter precursors, metabolic substrates, trophic factors and cytokines. Thus, astrocytes regulate the synaptic connectivity of the neuronal networks by supporting neurotransmitter metabolism, synaptogenesis, synaptic elimination and synaptic plasticity contributing to cognitive processing including learning, memory, emotions and behaviour. Astroglia-specific regulatory pathways affect the most fundamental properties of neuronal networks from their excitability to synaptic connectivity. Thus, it is the concerted action of glia and neurones, which, by employing distinct mechanisms, produce behavioural outputs of the ultimate control centre that we call the brain.

Brain Informatics and Health, 2016

The past few years have seen new research methods confirming more confidently that glia have a key information processing role in the brain, specifically in relation to learning capability. However, many details Tof glia's role remain unknown, including a gap between cellular and behavioural level findings. Based on Ca 2+ wave mechanics in astrocytes, we derive a theoretical capability of astrocytes to encode cognitive representations as probability distributions over synapses. The process is analogous to MCMC Bayesian inference that samples a neural network configuration from a prior in the astrocyte and then uses its performance to update to a posterior distribution. The proposed model explains recent behavioural results where obstructing astrocytes leads to deficiencies in learning new knowledge without affecting ability to recall existing knowledge. The model is also a novel Bayesian brain theory which uniquely addresses the cellular and synaptic levels.