Hippocampal representations of foraging trajectories depend upon spatial context

Area of Science:

• Neuroscience
• Behavioral Neuroscience
• Cognitive Science

Background:

• Animals learn reward trajectories in both spatial and non-navigational contexts.
• Synchronous hippocampal activity is crucial for recalling and evaluating learned trajectories.

Purpose of the Study:

• To investigate if hippocampal representations differentially contribute to experience-dependent learning of trajectories across spatial and non-navigational contexts.
• To understand the role of hippocampal activity in reinforcement learning across different task types.

Main Methods:

• Mice were trained on two tasks: navigating a physical arena to a hidden target and manipulating a joystick to a virtual target for delayed rewards.
• Calcium imaging was used in freely moving mice to record hippocampal (CA1) activity during task performance.

Main Results:

• In the navigational context, synchronous CA1 reactivation was retrospective, aiding the evaluation of prior trajectories.
• In the non-navigational context, CA1 reactivation was prospective, facilitating the initiation of joystick trajectories.
• Trajectory adaptation to new targets was explained by a common learning algorithm with context-dependent hippocampal contributions.

Conclusions:

• Hippocampal representations play distinct roles in learning trajectories depending on the spatial context.
• Synchronous CA1 reactivation differentially supports retrospective evaluation in navigation and prospective initiation in non-navigational tasks.
• These findings suggest a flexible role for the hippocampus in reinforcement learning across diverse contexts.