Within brain tissue, astrocytes represent by far the most abundant cell lineage. It has become widely accepted that astrocytes are anything but the glue of the Central Nervous System and are associated, among others, with cognitive functions, information flow and processing, metabolic regulation and even the pathogenesis of certain neurodegenerative diseases. Astrocytes establish network arrangements, both unique to them (pure astrocytes network -PAN-) and mingled with other neural cells (neural network -NN-).

A puzzling question is how the coupling strength of the assembly elements affects the physiological behavior of any individual astrocytes within the network and vice versa. No methodology is, as yet, available to tackle these challenging questions

The present work proposes to study astrocyte networks from an interdisciplinary approach, identifying the system structure that determines a particular property  (i.e. connectivity) and the temporal dynamics that characterize the baseline behaviour of the system and in response to specific disturbances.

To this end we define a 2D Cartesian lattice of interconnected nodes with first neighbors connectivity where the nodes can be interpreted in terms of the cell forming the network. Our proposal is to share a model of automata as a heuristic playground for the biologist. A model robust, comprehensive, descriptive and predictive enough to test possible hypotheses from a set of experimental data.

We implement various scenarios and alternative rules for the interaction between network elements: e.g. the number of responding elements and the weight of such interactions, the driving force of the exchange, the spatio-temporal propagation of the response and variable and constant exchange coefficients.

The incorporation of parameters on which the system's behaviour depends allows us to build a more comprehensive model that renews the repertoire of questions to ask about the homeostasis of the system. In this stage of the work, we present the main strengths and problems detected with our simulated scenarios