Distributed processing in cultured neuronal networks.

Thanks to a number of recent technical advances, it will become increasingly popular to study the very basics of distributed information processing using cultured neuronal networks. Most researchers studying population coding are working with intact, living animals. Clearly, cultured neuronal networks lack many features of real brains, but they retain many others. They develop organotypic synaptic connections and exhibit a rich variety of distributed patterns of electrical activity. Progress in multi-electrode array technology, optical recording, and multi-photon microscopy, has made it possible that every cell in a cultured monolayer network can be observed, monitored, stimulated, and manipulated with temporal resolution in the submillisecond range, and spatial resolution in the submicron range, in a non-destructive manner. At present, such detailed and complete analysis of neural circuits is not feasible in living animals, or even brain slices. It is an open question, however, whether any of the ‘processing’ done by cultured neurons is relevant to that carried out by intact brains. This chapter serves to present efforts from a number of groups that lay the groundwork for an in vitro approach to studying population coding. I will suggest what it might take to advance the state of the art to the point where we can consider studying learning, memory, and distributed information processing in vitro.