Abstract:Hippocampal circuits in the brain enable two distinct cognitive functions: the construction of spatial maps for navigation and the storage of sequential episodic memories. This dual role remains an enduring enigma. While there have been advances in modeling spatial representations in the hippocampus, we lack good models of its role in episodic memory. Here we present a neocortical-entorhinal-hippocampal network model that implements a high-capacity general associative memory, spatial memory, and episodic memory by factorizing content storage from the dynamics of generating error-correcting stable states. Unlike existing neural memory models, which exhibit a memory cliff, the circuit (which we call Vector-HaSH, Vector Hippocampal Scaffolded Heteroassociative Memory) exhibits a graceful tradeoff between number of stored items and detail. Next, we show that structured internal scaffold states are essential for constructing episodic memory: they enable high-capacity sequence memorization by abstracting the chaining problem into one of learning low-dimensional transitions. Finally, we show that previously learned spatial sequences in the form of cortico-hippocampal location-landmark associations can in turn be used as larger scaffolds and associated with neocortical inputs for a high-fidelity one-shot memory, providing the first circuit model of the "memory palaces" used in the striking feats of memory athletes.

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is about how the hippocampus simultaneously supports two different cognitive functions in the brain: constructing spatial maps for navigation and storing sequential episodic memories. Specifically, the paper explores the following points: 1. **The dual role of the hippocampus**: The hippocampus is not only involved in the formation of spatial memory, but also responsible for storing episodic memories of personal experiences. Why these two forms of memory co - exist in the same structure has always been a mystery in neuroscience. 2. **Limitations of existing models**: Existing neural memory models usually exhibit the "memory cliff" phenomenon when dealing with a large number of stored items, that is, after exceeding a certain number, the memory performance drops sharply. The paper proposes a new model - Vector - HaSH (Vector Hippocampal Scaffolded Heteroassociative Memory), aiming to overcome this limitation. 3. **Key mechanisms of the model**: - **Separation of content storage and dynamic generation**: The Vector - HaSH model achieves high - capacity storage by separating the content storage from the process of dynamically generating fixed points. Fixed points are generated by pre - structured grid cell circuits and interact with the hippocampus through randomly fixed connections, while the content is stored through the hetero - association of these fixed points with input data. - **Low - dimensional transformation operation**: The model maps the transformation dynamics in episodic memory onto low - dimensional grid states, thereby achieving efficient sequential memory. 4. **Circuit model of the memory palace technique**: The paper further shows that previously learned spatial sequences can serve as a larger scaffold, associated with neocortical input, to achieve high - fidelity single - trial memory, which is the first circuit model of the "memory palace" technique used by memory athletes. 5. **Theoretical and empirical analysis**: The paper discusses in detail the theoretical and empirical properties of this model, including its performance in different settings, from spatial memory to non - spatial memory and then to sequential episodic memory. Through these studies, the paper aims to provide a more comprehensive and effective model to explain how the hippocampus efficiently performs multiple memory functions with limited resources.

Episodic and associative memory from spatial scaffolds in the hippocampus

A hippocampus CA3 spiking neural network model for storage and retrieval of sequential memory

Associative Memory Model of Hippocampus CA3 Using Spike Response Neurons

How the Brain Formulates Memory: A Spatio-Temporal Model Research Frontier.

The hippocampus, memory, and spatial function

A generative model of memory construction and consolidation

Topological Schemas of Memory Spaces

Neural models of episodic memory

A Computational Theory Of Episodic Memory Formation In The Hippocampus

A Computationally Efficient Associative Memory Model of Hippocampus CA3 by Spiking Neurons

A hierarchy of associations in hippocampo-cortical systems: cognitive maps and navigation strategies

Spiking neural network-based computational modeling of episodic memory

A Model Of The Operation Of The Hippocampus And Entorhinal Cortex In Memory

The Storage and Recall of Memories in the Hippocampo-Cortical System

A Hierarchical Organized Memory Model Using Spiking Neurons

Episodic Memory formation: A review of complex Hippocampus input pathways

Scale-Dependent Coding of the Hippocampus in Relational Memory

Memory and Space: Towards an Understanding of the Cognitive Map

Self-organization of Multiple Spatial and Context Memories in the Hippocampus

Extensive Cortical Connectivity of the Human Hippocampal Memory System: Beyond the “what” and “where” Dual Stream Model

Cortical reactivation of spatial and non-spatial features coordinates with hippocampus to form a memory dialogue