Abstract
Animals adapt their behavior in response to changing environments, a process that requires cognitive flexibility mediated by coordinated activity across multiple brain regions. In the context of reward-guided spatial memory and behavior, the hippocampus encodes a cognitive map, the prefrontal cortex (PFC) is crucial for rule representation and executive control; and dopaminergic (DA) neurons in the ventral tegmental area (VTA) convey reward-related signals. To better understand how information processing in these regions evolves with changing reward conditions, we trained rats on a rule-switching task while performing high-density electrophysiological recordings from dorsal CA1, PFC and VTA. Single cell spiking activity in CA1 and PFC was modulated by task rules. Rule decoding using their population activity closely aligned with animal’s behavior strategy. VTA DA neurons exhibited reward-predictive activity during running, a feature that emerged during behavior adjustments, and peaked as PFC rule decoding began to reflect the new rule. Notably, changes in neural decoding and reward prediction preceded behavioral strategy transitions by several trials. We further investigated possible mechanisms underlying coordination between these brain regions. PFC-VTA theta coherence increased upon reward delivery, in comparison to reward omission. DA neuron firing activity was modulated during hippocampal sharp-wave ripples (SWRs). Co-modulated pairs were more prevalent between VTA DA neurons and rule-modulated CA1 neurons. In addition, CA1-DA spiking relationships varied dynamically across different stages of behavioral performance.
Together, these findings revealed temporal coordination between neural coding and behavioral adaptation during rule switching, and implied cross-regional communication through transiently synchronized activity during both online and offline processing.