Sequence learning induces stimulus specific changes in resting-state brain connectivity on multiple timescales

Abstract

Repeated reactivation of recently acquired information has been put forward as a candidate mechanism supporting long-term memory consolidation. Evidence from rodent electrophysiology studies suggests that the hippocampus plays a central role in coordinating the reactivation of neocortical memory representations during post-encoding offline periods. However, evidence from human neuroimaging studies is scarce and it remains unclear at which timescales offline memory reactivation occurs. In the current study, we used functional magnetic resonance imaging (fMRI) and magnetoencephalography (MEG) to investigate changes in the resting-state brain connectivity pattern following a sequence learning task. Participants were required to remember several multi-item sequences, comprising alternating face and scene or faceless body stimuli. Preliminarily results suggest learning-induced changes in fMRI and source space MEG resting-state connectivity when comparing post-encoding rest with preexperimental baseline periods. Specifically, we observe increased correlation between the hippocampus and specialized ventral visual regions processing the categorical sequence stimuli. Moreover, the stimulus-specific connectivity pattern across regions during encoding appears to show increased similarity to the connectivity pattern observed during post-encoding relative to the baseline rest period. These preliminary findings are in line with evidence suggesting hippocampus mediated neocortical reactivation of recently acquired information during subsequent offline periods. By comparing learning-induced changes in resting-state connectivity from different neuroimaging modalities and relating these observations to subsequent memory performance, we aim at identifying network signatures of memory reactivation at different timescales.