Since the 1960s, the large neurones of some invertebrates have been exploited in attempts to define the neural circuits that underlie simple behaviours. Even in the relatively 'simple' nervous systems of these animals, it is often difficult to study individual synaptic connections in detail and to rule out involvement of unidentified neurones. These limitations have been overcome by reconstruction of partial circuits of identified neurones in cell culture. This approach has provided opportunities to examine the function of small neuronal circuits in a manner that is unapproachable in the intact nervous system.