Glider compositionality and information capacity in an E/I-balanced neuronal automaton

Near excitation/inhibition balance in an E/I-balanced neuronal automaton, sparse interacting gliders coexist with high computational capacity, with local capacity varying across space and time.

• Why it matters: It remains unclear whether neuronal network models can exploit the information-carrying and transforming behavior of gliders seen in cellular automata.
• What they did: The authors simulate a cortical network model with excitatory and inhibitory populations, graded cellular activity, and uniform local connectivity, then use an analytical reduction to identify conditions that sustain the dominant G-glider.
• The result: Distinct gliders assemble into growing structures that emit other glider types, and input differentiation/generalization are high near balance but low under strong interneuron mutual inhibition, suggesting interacting gliders as a mechanism for dynamic plasticity.

AI summary of the abstract · may contain errors · not medical advice · read the paper.