Hippocampo-neocortical interaction as compressive retrieval-augmented generation
- 1Nuffield Department of Clinical Neurosciences, University of Oxford, Oxford, UK. [email protected]
- 2Institute of Cognitive Neuroscience, University College London, London, UK. [email protected]
- 3Queen Square Institute of Neurology, University College London, London, UK. [email protected]
Abstract
Many aspects of learning, memory, and problem solving involve interplay between episodic (hippocampal) and semantic (neocortical) systems, but the neural mechanisms supporting this are unclear. We present a computational model in which sequential experiences are encoded in hippocampus in compressed form and replayed to train a neocortical generative network. This network captures the gist of specific episodes and extracts statistical patterns that generalise to new situations, enabling efficient reconstruction of the past and prediction of the future. The two systems interact during encoding, recall and problem solving, with the hippocampus retrieving relevant episodic information into working memory as a basis for generation using the 'general knowledge' of the neocortical network. We simulate this interaction as 'retrieval-augmented generation', with the addition of mechanisms to compress episodic memories into hippocampus and to consolidate them into neocortex. The model explains changes to memories over time, including schema-based distortions, and shows how episodic and semantic memory contribute to problem solving.
Related
- Next generation Human Neocortical Neurosolver model with biophysically realistic somatic, dendritic and synaptic dynamics
- Hippocampo-septal inhibitory feedback controls the medial septal circuit as a function of brain state, pyramidal cell activity and movement speed.
- RAGCell: Retrieval-Augmented Generation as Supervision for Versatile Single-cell Analysis
- An organotypic hippocampo-prefrontal co-culture model for studying long-range connectivity and circuit plasticity
- RiboBridge couples RNA backbone generation to sequence design through a learned interaction topology
BibTeX
@article{spens2026hippocamponeocortical,
title = {{Hippocampo-neocortical interaction as compressive retrieval-augmented generation}},
author = {Eleanor Spens and Neil Burgess},
journal = {Nature Communications},
year = {2026},
month = jun,
doi = {10.1038/s41467-026-74357-6},
url = {https://doi.org/10.1038/s41467-026-74357-6},
note = {CC BY},
}
RIS
TY - JOUR
TI - Hippocampo-neocortical interaction as compressive retrieval-augmented generation
AU - Eleanor Spens
AU - Neil Burgess
PY - 2026
DA - 2026/06/25
JO - Nature Communications
DO - 10.1038/s41467-026-74357-6
UR - https://doi.org/10.1038/s41467-026-74357-6
AB - Many aspects of learning, memory, and problem solving involve interplay between episodic (hippocampal) and semantic (neocortical) systems, but the neural mechanisms supporting this are unclear. We present a computational model in which sequential experiences are encoded in hippocampus in compressed form and replayed to train a neocortical generative network. This network captures the gist of specific episodes and extracts statistical patterns that generalise to new situations, enabling efficient reconstruction of the past and prediction of the future. The two systems interact during encoding, recall and problem solving, with the hippocampus retrieving relevant episodic information into working memory as a basis for generation using the 'general knowledge' of the neocortical network. We simulate this interaction as 'retrieval-augmented generation', with the addition of mechanisms to compress episodic memories into hippocampus and to consolidate them into neocortex. The model explains changes to memories over time, including schema-based distortions, and shows how episodic and semantic memory contribute to problem solving.
SN - 2041-1723
ER -