A neural circuit for flexible control of persistent behavioral states

To adapt to their environments, animals must generate behaviors that are closely tuned to a dynamically changing sensory world. However, behavioral states such as foraging or mating typically persist over long time scales to ensure proper execution. It remains unclear how neural circuits generate stable activity patterns to drive behavioral states, while maintaining the flexibility to select among alternative states when the sensory context changes. Here, we elucidate the functional architecture of a neural circuit controlling the choice between exploration and exploitation states during foraging in C. elegans. We identify stable circuit-wide activity patterns underlying each behavioral state and show that feedback between a sensorimotor circuit and two antagonistic neuromodulatory inputs underlies the emergence of these network states. Sensory processing neurons that detect salient food cues can couple to either neuromodulatory system and bias the network towards different states in different sensory contexts. This allows animals to dynamically adjust the balance between exploration and exploitation during foraging. Our results demonstrate that bi-directional communication between sensorimotor and neuromodulatory circuits allows animals to flexibly select behavioral states appropriate for their sensory context. This neural circuit motif may be broadly used in a variety of cases where animals need to balance behavioral persistence with sensitivity to environmental change.