Wednesday, March 12, 2025
Home Neuroscience New study links spatial navigation and language processing in the brain

New study links spatial navigation and language processing in the brain

by Medical Xpress
0 comment


DSI representations for the space. (A) The square room tiled with 30 × 30 discrete states. (B) Grid-like spatial representations generated by DSI-decorr (Upper) and their spatial autocorrelation (Lower). (C) Spatial representations like place cells generated by DSI-sparse. Credit: Proceedings of the National Academy of Sciences (2025). DOI: 10.1073/pnas.2413449122

Scientists at the Okinawa Institute of Science and Technology (OIST), the National Institute of Information and Communications Technology, and the University of Tokyo have found a mathematical connection between spatial navigation and language processing, creating a model called “Disentangled Successor Information” (DSI).

This model generates patterns that closely resemble the activity of actual brain cells involved in both spatial awareness (place cells and grid cells) and concept recognition (concept cells).

The DSI model shows that the hippocampus and entorhinal cortex— previously known primarily for —likely use comparable computational processes to handle both physical spaces and meaningful ideas or words. Using this shared framework, both types of information can be processed through similar mathematical computations, which could be achieved in the brain by partial activation of specific groups of neurons.

This discovery shows that how we navigate spaces and how we understand language might use the same basic brain processes. The findings have been published in PNAS.

New study links spatial navigation and language processing in the brain
DSI representations capture the semantic structure of words at the population level (DSI-decorr). (A) The rank correlation of word similarity evaluated by word representation vectors (cosine similarity) and humans (WS353 dataset). For DSI, dots indicate 5 trials with different random seeds (different initial values for learning); bars indicate means of those 5 simulations. (B) Dissimilarity matrix between DSI representation vectors for 100 words in 10 semantic categories (DSI-decorr). We selected 10 words in each category. We used same dissimilarity metric with Reber et al. (32) (1 − Pearson’s correlation coefficient). (C) Visualization of the representational structure of DSI using MDS based on the dissimilarity matrix shown in (B). Each color corresponds to a semantic category. Credit: Proceedings of the National Academy of Sciences (2025). DOI: 10.1073/pnas.2413449122

More information:
Tatsuya Haga et al, A unified neural representation model for spatial and conceptual computations, Proceedings of the National Academy of Sciences (2025). DOI: 10.1073/pnas.2413449122

Citation:
New study links spatial navigation and language processing in the brain (2025, March 11)
retrieved 11 March 2025
from https://medicalxpress.com/news/2025-03-links-spatial-language-brain.html

This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no
part may be reproduced without the written permission. The content is provided for information purposes only.





Source link

You may also like

Leave a Comment

Edtior's Picks

Latest Articles

All Rights reserved, site designed by Yellohost.co.za