Frontiers in Psychiatry (Feb 2020)

The Neural Mechanisms of Associative Memory Revisited: fMRI Evidence from Implicit Contingency Learning

  • Marco P. Caviezel,
  • Carolin F. Reichert,
  • Carolin F. Reichert,
  • Dena Sadeghi Bahmani,
  • Dena Sadeghi Bahmani,
  • Dena Sadeghi Bahmani,
  • Christoph Linnemann,
  • Caroline Liechti,
  • Caroline Liechti,
  • Oliver Bieri,
  • Stefan Borgwardt,
  • Stefan Borgwardt,
  • Thomas Leyhe,
  • Thomas Leyhe,
  • Tobias Melcher

DOI
https://doi.org/10.3389/fpsyt.2019.01002
Journal volume & issue
Vol. 10

Abstract

Read online

The literature describes a basic neurofunctional antagonism between episodic memory encoding and retrieval with opposed patterns of neural activation and deactivation, particularly in posterior midline regions. This has been coined the encoding/retrieval (E/R) flip. The present fMRI study uses an innovative task paradigm to further elucidate neurofunctional relations of encoding and retrieval in associative memory. Thereby, memory encoding is implemented as implicit (non-deliberate) cognitive process, whereas the prior literature focused mainly on explicit encoding. Moreover, instead of defining brain activations related to successful (vs. unsuccessful) memory performance, the task paradigm provides proper no-memory baseline conditions. More specifically, the encoding task includes trials with non-contingent (not learnable) stimulus combinations, while the retrieval task uses trials with a simple matching exercise with no mnemonic requirements. The analyses revealed circumscribed activation in the posterior middle cingulate cortex (pMCC) together with prominent deactivation in the anterior insula cortex (aIC) as core neural substrate of implicit memory encoding. Thereby, the pMCC exhibited positive functional connectivity to the hippocampus. Memory retrieval was related to an activation pattern exactly opposed to memory encoding with deactivation in the pMCC and activation in the aIC, while the aIC additionally exhibited a negative (i.e., arguably inhibitive) functional connectivity to the pMCC. Important to note, the observed pattern of activations/de-activations in the pMCC appears to conflict with prevalent E/R flip findings. The outlined results and their (alleged) discrepancies with prior study reports are discussed primarily in the context of the default mode network’s functioning and its context-sensitive regulation. Finally, we point out the relevance of the present work for the understanding and further investigation of the neurofunctional aberrations occurring during normal and pathological aging.

Keywords