A Computational Model of Systems Memory Reconsolidation and Extinction

In the mammalian brain newly acquired memories are dependent on the hippocampus for maintenance and recall, but over time these functions are taken over by the neocortex through a process called memory consolidation. Thus, whereas recent memories are likely to be disrupted in the event of hippocampal damage, older memories are less vulnerable. However, if a consolidated memory is reactivated by a reminder, it can temporarily return to a hippocampusdependent state. This phenomenon is known as memory reconsolidation. Here, we present an artificial neural-network model that captures memory consolidation and reconsolidation, as well as the related phenomena of memory extinction, spontaneous recovery and trace dominance. The model provides a novel explanation of trace dominance, the competition between reconsolidation and extinction, from which we derive two predictions that could be tested in future experiments.