Remote memory in a Bayesian model of context fear conditioning (BaconREM)

Area of Science:

• Neuroscience
• Cognitive Science
• Computational Psychiatry

Background:

• Context fear learning is crucial for survival.
• Remote memories typically become independent of the hippocampus.
• Existing models explain recent fear but not remote fear phenomena.

Purpose of the Study:

• To propose a computational model (BaconREM) of remote context fear memory.
• To simulate and predict key aspects of remote fear memory consolidation.
• To account for hippocampus-independence and generalization changes over time.

Main Methods:

• Extension of the Bayesian overlapping network (BACON) model.
• Computational simulation of rodent fear conditioning paradigms.
• Analysis of memory consolidation and generalization under different conditions.

Main Results:

• BaconREM predicts hippocampus-independence and hyper-generalization of remote fear.
• The model accounts for the effects of systems consolidation on memory.
• It also predicts that reminders can restore hippocampus-dependence and reduce generalization.

Conclusions:

• BaconREM offers a unified explanation for remote fear memory phenomena.
• The model suggests that remote contextual fear memory can reside in the neocortex.
• It provides a framework for understanding memory consolidation and reconsolidation.