The emergence of dynamical instantaneous memory in the spontaneous activity of spatially confined neuronal assemblies in vitro

Abstract

Understanding the dynamics between communicating cell assemblies is essential for deciphering the neural code and identifying the mechanism underlying memory formation. In this work, in order to unveil possible emergent intrinsic memory phenomena in the communication between cell assemblies, we study the spontaneous dynamics of in vitro spatially confined inter-connected neuronal circuits grown on multi-electrode arrays. The spontaneous dynamics of the global network was characterized by the coupling of the activity independently generated by each circuit.The asymptotic functional connectivity of the network reflected its modular organization. Instantaneous functional connectivity maps on ten seconds epochs, revealed more complex dynamical states with the simultaneous activation of distinct circuits. When looking at the similarity of the generated network events, we observed 2 that spontaneous network events occurring at temporal distances below two dozens of seconds had an average higher similarity compared to randomly played network events. Such a memory phenomenon was not observed in networks where spontaneous events were less frequent and in networks topologically organized as open lines. These results support the hypothesis that dynamical instantaneous memory, characterized by drifting network dynamics with decaying degree of similarity, is an intrinsic property of neuronal networks.

Experimental Platform to Study Spiking Pattern Propagation in Modular Networks In Vitro

“…The number of axons passing through microchannels regulates the transmission and the connectivity weight of neuronal modules [18][19][20]. Microelectrode arrays combined with microfluidic chips provide electrophysiological monitoring of spiking pattern generation and propagation within and between modular networks [20][21][22][23][24][25]. Such networks demonstrate a large number of spontaneous spiking patterns in contrast to uniform networks [26].…”

“…In cortical cultures, modularity enhances the classification of information, as represented by patterns of precise spike timings induced by local electrical stimulation [27]. Recently, a high similarity between close-in-time spontaneous activity patterns was observed in modular networks, suggesting instantaneous memory-like phenomena [24]. Here we developed and studied two types of microfluidic chips with asymmetric microchannels of various designs, which provide unidirectional connections between neuronal networks located in different chambers (modules).…”