Skip to content
DraconiticResearch
Search Sources API
189,105 studies31 sourcesUpdated 5 min ago
Peer-reviewedCC BY

Memory out of context: Spacing effects and decontextualization in a computational model of the medial temporal lobe

  • James W. Antony1,2,3,4
  • Xiaonan L. Liu5
  • Yicong Zheng2,4
  • Charan Ranganath A2,4
  • Randall C. O’Reilly2,4
5 affiliations
  1. 1Department of Psychology & Child Development, California Polytechnic State University
  2. 2Departments of Psychology and Computer Science, Center for Neuroscience, University of California
  3. 3Department of Psychology & Child Development, California Polytechnic State University, San Luis Obispo
  4. 4Departments of Psychology and Computer Science, Center for Neuroscience, University of California, Davis
  5. 5Department of Psychology, The Chinese University of Hong Kong
Open at sourcePDFDOICite

Abstract

Some neural representations change across multiple timescales. Here we argue that modeling this “drift” could help explain the spacing effect (the long-term benefit of distributed learning), whereby differences between stored and current temporal context activity patterns produce greater error-driven learning. We trained a neurobiologically realistic model of the entorhinal cortex and hippocampus to learn paired associates alongside temporal context vectors that drifted between learning episodes and/or before final retention intervals. In line with spacing effects, greater drift led to better model recall after longer retention intervals. Dissecting model mechanisms revealed that greater drift increased error-driven learning, strengthened weights in slower-drifting temporal context neurons (temporal abstraction), and improved direct cue-target associations (decontextualization). Intriguingly, these results suggest that decontextualization — generally ascribed only to the neocortex — can occur within the hippocampus itself. Altogether, our findings provide a mechanistic formalization for established learning concepts such as spacing effects and errors during learning.

Related

  1. A hierarchical memory architecture overcomes context limits in long-horizon multi-agent computational modeling
    Shivendra G. Tewari, Holly Kimko
    arXivPreprintarXiv licenceFull textBioinformatics & computational biologyarXiv
  2. Structure-based computational prioritization of Emodin-8-glucoside as a candidate ligand for the VEGFR1 kinase domain in the context of placenta previa
    Deniz Inan, Sinan Karageçili, Nouman Ali
    PLOS OnePeer-reviewedCC BY 4.0Full textCell biologyPharmacology & toxicologyPLOS
  3. A long neural circuit spanning multiple brain regions is involved in context-induced retrieval of fear memory
    Hao Yang, Ting Xia, Si-Ying Kuang +3
    Zoological ResearchPeer-reviewedCC BY-NC 4.0Full textNeurosciencePMC
  4. HIPPOCAMPAL TAU RELATED SIGNAL INDEXES HYPEREXCITABILITY AND NREM SLEEP MEMORY DYSFUNCTION IN TEMPORAL LOBE EPILEPSY
    T. Arafat, K. Schiller, J. Lam +21
    bioRxivPreprintAll rights reservedFull textNeurosciencebioRxiv
  5. A context-aware temporal modeling through unified multi-scale temporal encoding and hierarchical sequence learning for single-channel EEG sleep staging
    Amir Ali Vakili, Salar Jahanshiri, Armin Salimi-Badr
    PLOS OnePeer-reviewedCC BY 4.0Full textNeurosciencePhysiologyPLOS

Cite

BibTeXRIS
BibTeX
@article{antony2026memory,
  title = {{Memory out of context: Spacing effects and decontextualization in a computational model of the medial temporal lobe}},
  author = {James W. Antony and Xiaonan L. Liu and Yicong Zheng and Charan Ranganath A and Randall C. O’Reilly},
  year = {2026},
  month = jun,
  doi = {10.21428/8e6ba8ef.d9f5dd19},
  url = {https://doi.org/10.21428/8e6ba8ef.d9f5dd19},
  note = {CC BY},
}
RIS
TY  - JOUR
TI  - Memory out of context: Spacing effects and decontextualization in a computational model of the medial temporal lobe
AU  - James W. Antony
AU  - Xiaonan L. Liu
AU  - Yicong Zheng
AU  - Charan Ranganath A
AU  - Randall C. O’Reilly
PY  - 2026
DA  - 2026/06/15
DO  - 10.21428/8e6ba8ef.d9f5dd19
UR  - https://doi.org/10.21428/8e6ba8ef.d9f5dd19
AB  - Some neural representations change across multiple timescales. Here we argue that modeling this “drift” could help explain the spacing effect (the long-term benefit of distributed learning), whereby differences between stored and current temporal context activity patterns produce greater error-driven learning. We trained a neurobiologically realistic model of the entorhinal cortex and hippocampus to learn paired associates alongside temporal context vectors that drifted between learning episodes and/or before final retention intervals. In line with spacing effects, greater drift led to better model recall after longer retention intervals. Dissecting model mechanisms revealed that greater drift increased error-driven learning, strengthened weights in slower-drifting temporal context neurons (temporal abstraction), and improved direct cue-target associations (decontextualization). Intriguingly, these results suggest that decontextualization — generally ascribed only to the neocortex — can occur within the hippocampus itself. Altogether, our findings provide a mechanistic formalization for established learning concepts such as spacing effects and errors during learning.
ER  -