Converging assemblies: a putative building block for brain function and for interfacing with the brain

The organization of biological neuronal networks into functional modules has intrigued scientists and inspired engineers to develop artificial systems. These networks are characterized by two key properties. First, they exhibit dense interconnectivity (Braitenburg and Schüz, 1998; Campagnola et al., 2022). The strength and probability of connectivity depend on cell type, inter-neuronal distance, and species. Still, every cortical neuron receives input from thousands of other neurons while transmitting output to a similar number of neurons. Second, communication between neurons occurs primarily via chemical or electrical synapses. The transmission of information is mediated mainly during presynaptic spiking events that generate postsynaptic inward currents and intracellular depolarization, which in turn induce postsynaptic spikes. However, these two properties alone cannot explain the complex mechanisms of information processing in neuronal networks.