Oscillation-driven memory encoding, maintenance and recall in an entorhinal-hippocampal circuit model

Coherent neuronal activity phase-locked to theta and gamma oscillations is thought to be crucial for information processing across multiple brain regions. However, the network mechanisms underlying the oscillation-driven multi-area computation remains largely unclear. To explore such mechanisms, we constructed a hippocampal-entorhinal neural network model involving parvalbumin-positive, somatostatin-positive and vasoactive intestinal polypeptide interneurons in the hippocampal area CA1 and the superficial and deep layers of the entorhinal cortex. We examined the model behavior by using neural activity data recorded during delayed nonmatching to place tasks. Our model shows that experimentally observed relative phases of theta oscillation ensure working memory performance. Moreover, the model predicts that acetylcoline concentrations in these areas modulate the balance between intra-area and inter-area information processing according to cognitive demands emergent at encoding, maintenance and recall epochs of a working memory task. Our model suggests the active role of theta-phase-locked firing and cholinergic modulations for multi-area memory processing.