Recurrent neural networks as neuro-computational models of human speech recognition
PLOS Computational Biology · 2025
DOI: 10.1371/journal.pcbi.1013244 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS Computational Biology, 2025. Verfasst von Christian Brodbeck und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computing and Software, McMaster University, Hamilton, Ontario, Canada. Hauptthemen laut MEDLINE: Neural Networks, Computer, Speech Perception, Models, Neurological. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 72 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science Foundation, Basque Government, NSF.
Abstract (Originalsprache)
Human speech recognition transforms a continuous acoustic signal into categorical linguistic units, by aggregating information that is distributed in time. It has been suggested that this kind of information processing may be understood through the computations of a Recurrent Neural Network (RNN) that receives input frame by frame, linearly in time, but builds an incremental representation of this input through a continually evolving internal state. While RNNs can simulate several key behavioral observations about human speech and language processing, it is unknown whether RNNs also develop computational dynamics that resemble human neural speech processing . Here we show that the internal dynamics of long short-term memory (LSTM) RNNs, trained to recognize speech from auditory spectrograms, predict human neural population responses to the same stimuli, beyond predictions from auditory features. Variations in the RNN architecture motivated by cognitive principles further improved this predictive power. Specifically, modifications that allow more human-like phonetic competition also led to more human-like temporal dynamics. Overall, our results suggest that RNNs provide plausible computational models of the cortical processes supporting human speech recognition.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Department of Computing and Software, McMaster University, Hamilton, Ontario, Canada.
- Department of Psychological Sciences, University of Connecticut, Storrs, Connecticut, United States of America.
- Stellantis Group, Poissy, France.
- Computational Neuroscience, Basque Center on Cognition, Brain and Language, Donostia - San Sebastián, Spain.
- Ikerbasque: Basque Foundation for Science, Bilbao, Spain.
ORCID-Profile: Christian Brodbeck
Förderung
- National Science Foundation
- Basque Government
- NSF
- Spanish State Research Agency
Zitieren
APA
Brodbeck, C., Hannagan, T., & Magnuson, J. S. (2025). Recurrent neural networks as neuro-computational models of human speech recognition. PLOS Computational Biology. https://doi.org/10.1371/journal.pcbi.1013244
Vancouver
Brodbeck C, Hannagan T, Magnuson JS. Recurrent neural networks as neuro-computational models of human speech recognition. PLOS Computational Biology. 2025. doi:10.1371/journal.pcbi.1013244
BibTeX
@article{Brodbeck2025,
author = {Brodbeck, Christian and Hannagan, Thomas and Magnuson, James S},
title = {Recurrent neural networks as neuro-computational models of human speech recognition},
journal = {PLOS Computational Biology},
year = {2025},
doi = {10.1371/journal.pcbi.1013244}
}
EUVIMED