Unifying Space and Relational Memory

This piece, by Armin Bazarjani, was published on 11/19/24. The original text, by Whittington et al., was published by Cell on 11/25/20.

This study by Whittington et al. proposes a model that unifies two views of the hippocampus: spatial (Tolman) and non-spatial relational memory (Eichenbaum), which can both be unified under the common language of constructing useful generalizations. 

They propose medial entorhinal cells form a basis describing structural knowledge, and hippocampal cells link this basis with sensory representations. This is a very simple structuring of the problem. 

By factoring the common structural code away from the particular sensory code, you can reuse that same structural code for any particular sensory environment. They propose that relational memories bind the structural code to a particular sensory code for each environment. 

They show that this model can learn abstract concepts, like families, after only being told about certain familial relationships between certain individuals. 

Most inspirationally, this model has also been able to explain a wide range of neurological phenomena, including place cells, grid cells, border cells, object-vector cells, landmark cells, etc. Additionally, the model provides interesting explanations for hippocampal remapping, suggesting that it is not random but preserves structural knowledge across environments. 

This work is interesting and provides a compelling model for how spatial structure is tied to non-spatial relational memory.

References

Whittington, J. C., Muller, T. H., Mark, S., Chen, G., Barry, C., Burgess, N., & Behrens, T. E. (2020). The Tolman-Eichenbaum machine: unifying space and relational memory through generalization in the hippocampal formation. Cell, 183(5), 1249-1263.

Want to submit a piece? Or trying to write a piece and struggling? Check out the guides here!

Thank you for reading. Reminder: Byte Sized is open to everyone! Feel free to submit your piece. Please read the guides first though.

All submissions to berkan@usc.edu with the header “Byte Sized Submission” in Word Doc format please. Thank you!