Evidence for complementary learning systems within the hippocampus

Area of Science:

• Neuroscience
• Cognitive Science
• Computational Neuroscience

Background:

• The hippocampus is crucial for episodic memory, but also implicated in structure learning, which involves extracting regularities across experiences.
• Episodic memory and structure learning present conflicting demands: separation versus integration of experiences.
• The C-HORSE model proposes an anatomical division of labor within the hippocampus to resolve these conflicting demands.

Purpose of the Study:

• To evaluate the C-HORSE model's alignment with empirical findings on hippocampal structure learning.
• To explore the roles of the trisynaptic pathway (TSP) and monosynaptic pathway (MSP) in different cognitive functions.
• To synthesize evidence for functional dissociations across hippocampal pathways in relation to structure learning.

Main Methods:

• Review of empirical evidence from domains benefiting from regularity detection (statistical learning, motor sequence learning, inference, category learning).
• Assessment of the hippocampus's role in these domains and evidence for pathway-specific functions.
• Evaluation of findings in relation to the predictions of the C-HORSE model.

Main Results:

• Strong convergence between the C-HORSE model and existing empirical literature on hippocampal function.
• Evidence suggests the monosynaptic pathway (MSP) plays a key role in structure learning.
• The trisynaptic pathway (TSP) is associated with rapid learning and pattern separation for episodic memory.

Conclusions:

• The C-HORSE model provides a coherent framework for understanding hippocampal contributions to both episodic memory and structure learning.
• The monosynaptic pathway (MSP) is identified as a critical component for extracting regularities and supporting prediction and generalization.
• Further experimental and modeling work is needed to refine our understanding of hippocampal pathway specialization.