Distinct spatial maps and multiple object codes in the lateral entorhinal cortex

Huang, Xu; Schlesiger, Magdalene I.; Barriuso-Ortega, Isabel; Leibold, Christian; MacLaren, Duncan A. A.; Bieber, Nina; Monyer, Hannah

doi:10.1016/j.neuron.2023.06.020

Cited by 8 publications

(4 citation statements)

References 77 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We found that we can apply the same computational framework to spatial representations. Furthermore, our results suggest a novel computational role of non-grid (heterogenous) spatial representations that have been experimentally found in EC 34,50 .…”

Section: Introductionsupporting

confidence: 65%

“…Furthermore, our model suggests the importance of non-grid spatial representations for inference (or switching) of spatial contexts. In EC, there are many neurons that exhibit non-grid spatial representations, and they show large remapping across contexts (34, 50). Previous models have shown that non-grid (heterogenous) spatial representations in EC support precise spatial encoding (36) and reward-based modulation of path integration (40).…”

Section: Discussionmentioning

confidence: 99%

“…Our results give alternative explanation of such remapping: such remapping can support analogical inference of navigation strategy across spatial contexts. However, it is still to be investigated whether this computational framework extends to general contextual changes such as wall colors, odors and reward settings 34,40,50 . We speculate that this is possible if we can appropriately construct disentangled representations for general sensory and task variables 53,62,63 .…”

Section: Theory Of Grid and Non-grid Representations In Ecmentioning

confidence: 99%

“…However, theoretical understanding of the relationship between spatial and conceptual representations has been scarce, especially for highly complex conceptual structures like languages. Although several models have recently worked on this problem (46)(47)(48)(49), it is often difficult to discuss computational properties of neural representations other than place cells and grid cells, such as concept cells (25)(26)(27) and non-grid spatial representations that represent contextual information in EC (34,50). Our model reveals a strong theoretical connection between spatial and semantic computations in the brain, which are biologically interpretable at unit-level and at population-level (Figure 1).…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A unified neural representation model for spatial and semantic computations

Haga

Oseki

Fukai

2023

Preprint

View full text Add to dashboard Cite

The entorhinal cortex uses grid-like neural representations (grid cells) for spatial processing. Neural representations for concepts (concept cells), which presumably support high-level conceptual processing, have been also found in the same brain region. However, the relationship behind those neural representations is still unclear. We propose a unified model called "disentangled successor information (DSI)" that produces neural representations for physical space and linguistic concepts. DSI generates grid-like representations in a 2-dimensional space that highly resemble those observed in the brain. Moreover, the same model creates concept-specific representations from linguistic inputs, corresponding to concept cells. Mathematically, DSI approximate value functions for navigation and word vectors obtained by word embedding methods, thus those representations support both spatial navigation and analogical inference based on a simple arithmetic computation. Our results suggest that representations and computation for physical space and linguistic concepts in the brain can emerge from a shared mechanism.

show abstract

Section: Introductionsupporting

confidence: 65%