Learning to combine top-down context and feed-forward representations under ambiguity with apical and basal dendrites
Cerebral Cortex · 2025
DOI: 10.1093/cercor/bhaf134 ↗Kurz gefasst
Fachartikel in der Zeitschrift Cerebral Cortex, 2025. Verfasst von Nizar Islah und 5 weiteren Autor:innen. Beteiligte Einrichtung: Centre de Recherche Azrieli du CHU Ste-Justine, Université de Montréal, 3175 Chem. de la Côte-Sainte-Catherine, Montréal H3T 1C5, Quebec, Canada. Hauptthemen laut MEDLINE: Pyramidal Cells, Dendrites, Neural Networks, Computer, Models, Neurological. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 65 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Institute for Data Valorization, Unifying Neuroscience and Artificial Intelligence-Québec (UNIQUE), CHU Sainte-Justine Azrieli Research Center (CHUSJ-ARC).
Abstract (Originalsprache)
One of the hallmark features of neocortical anatomy is the presence of extensive top-down projections into primary sensory areas. It is hypothesized that one of the roles of these top-down projections is to carry contextual information that helps animals to resolve ambiguities in sensory data. One proposed mechanism of contextual integration is a combination of input streams at distinct apical and basal dendrites of pyramidal neurons. Computationally, however, it is yet to be demonstrated how such an architecture could leverage distinct compartments for flexible contextual integration and sensory processing. Here, we implement a deep neural network with distinct apical and basal compartments that integrates (a) contextual information from top-down projections to apical compartments and (b) sensory representations driven by bottom-up projections to basal compartments. In addition, we develop a new contextual integration task using generative modeling. The performance of deep neural networks augmented with our “apical prior” exceeds that of single-compartment networks. We find that a sparse subset of neurons of the context-relevant categories receive the largest top-down signals. We further show that this sparse gain modulation is necessary. Altogether, this suggests that the “apical prior” could be key for handling the ambiguities that animals encounter in the real world.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Centre de Recherche Azrieli du CHU Ste-Justine, Université de Montréal, 3175 Chem. de la Côte-Sainte-Catherine, Montréal H3T 1C5, Quebec, Canada.
- Département d'informatique et de recherche opérationnelle, Université de Montréal, Pavillon André-Aisenstadt, 2920, chemin de la Tour local 2194, Montréal, H3T 1N8, Québec, Canada.
- Mila - Quebec AI Institute, 6666 Rue Saint-Urbain, Montréal H2S 3H1, Quebec, Canada.
- Department of Neurology and Neurosurgery, McGill University, 1033 Pine Avenue West, Montréal H3A 1A1, Quebec, Canada.
- Montreal Neurological Institute, McGill University, 3801 Rue University, Montréal H3A 2B4, Quebec, Canada.
- Department of Quantitative Life Sciences, McGill University, 550 Sherbrooke W., Montréal H3A 1E3, Quebec, Canada.
Förderung
- Institute for Data Valorization
- Unifying Neuroscience and Artificial Intelligence-Québec (UNIQUE)
- CHU Sainte-Justine Azrieli Research Center (CHUSJ-ARC)
- Fonds de Recherche du Québec-Santé
- Canadian Institute for Advanced Research (CIFAR) AI Chairs Program
- Quebec Institute for Artificial Intelligence (Mila)
- Natural Sciences and Engineering Research Council
Zitieren
APA
Islah, N., Etter, G., Tugsbayar, M., Gurbuz, B. T., Richards, B., & Muller, E. B. (2025). Learning to combine top-down context and feed-forward representations under ambiguity with apical and basal dendrites. Cerebral Cortex. https://doi.org/10.1093/cercor/bhaf134
Vancouver
Islah N, Etter G, Tugsbayar M, Gurbuz BT, Richards B, Muller EB. Learning to combine top-down context and feed-forward representations under ambiguity with apical and basal dendrites. Cerebral Cortex. 2025. doi:10.1093/cercor/bhaf134
BibTeX
@article{Islah2025,
author = {Islah, Nizar and Etter, Guillaume and Tugsbayar, Mashbayar and Gurbuz, Busra Tugce and Richards, Blake and Muller, Eilif B},
title = {Learning to combine top-down context and feed-forward representations under ambiguity with apical and basal dendrites},
journal = {Cerebral Cortex},
year = {2025},
doi = {10.1093/cercor/bhaf134}
}
EUVIMED