A Neural-Astrocytic Network Architecture

Abstract

Understanding the role of astrocytes in brain computation is a nascent challenge, promising immense rewards, in terms of new neurobiological knowledge that can be translated into artificial intelligence. In our ongoing effort to identify principles endowing the astrocyte with unique functions in brain computation, and translate them into neural-astrocytic networks (NANs), we propose a biophysically realistic model of an astrocyte that preserves the experimentally observed spatial allocation of its distinct subcellular compartments. We show how our model may encode, and modulate, the extent of synchronous neural activity via calcium waves that propagate intracellularly across the astrocytic compartments. This relationship between neural activity and astrocytic calcium waves has long been speculated but it is still lacking a mechanistic explanation. Our model suggests an astrocytic "calcium cascade" mechanism for neuronal synchronization, which may empower NANs by imposing periodic neural modulation known to reduce coding errors. By expanding our notions of information processing in astrocytes, our work aims to solidify a computational role for non-neuronal cells and incorporate them into artificial networks. CCS CONCEPTS• Computing methodologies → Model verification and validation; • Applied computing → Biological networks; Systems biology; Bioinformatics; KEYWORDS Neural Astrocytic Networks (NANs), Brain-morphic Architecture, Non-Von Neumann computing model, Event-based system ACM Reference Format: Ioannis Polykretis, Vladimir Ivanov, and Konstantinos P. Michmizos. 2018. A Neural-Astrocytic Network Architecture: Astrocytic calcium waves modulate synchronous neuronal activity.

Gridbot

“…To be effective in real-world, an autonomous agent, either biological or artificial, should 1) be robust to a noisy neural representation, 2) adapt to a fast-changing environment, 3) learn with no or limited supervision or reinforcement, and 4) compute efficiently with resource limitations. Our biologically constrained SNN overcomes the main problem that SNN architectures have, that of slow or computationally inefficient learning, and paves the way for introducing non-neuronal cells, such as astrocytes, that also process and learn information in the brain [38,39,55]. The application of SNNs in controlling a behavior, such as a motor task has indeed been impeded mainly by the lack of efficient or biologicallyconstrained learning methods [5,13,15,64].…”

Astrocytic modulation of neuronal signalling

“…Interestingly, a study employed neuronal cultures grown on a multielectrode array to explore the mechanism and origin of synchronised burst (SB) activity along with the application of computational modelling and has implicated the role of astrocytes in the generation reverberating activities in an SB ( Huang et al, 2017 ). Similarly, astrocytic Ca 2+ is predicted to be involved in the synchronised activity of large-scale neuronal ensembles ( Li et al, 2016b ; Polykretis et al, 2018 ) and can also cause intermittent neuron synchrony via slow astrocytic Ca 2+ oscillations ( Makovkin et al, 2020 ). In summary, both experimental and simulation studies suggest that astrocytes play a vital role in neuronal synchronization.…”