Interpretable functional specialization emerges in deep convolutional networks trained on brain signals
Abstract: Objective. Functional specialization is fundamental to neural information processing. Here, we study whether and how functional specialization emerges in artificial deep convolutional neural networks (CNNs) during a brain–computer interfacing (BCI) task. Approach. We trained CNNs to predict hand movement speed from intracranial electroencephalography (iEEG) and delineated how units across the different CNN hidden layers learned to represent the iEEG signal. Main results. We show that distinct, functionally interpretable neural populations emerged as a result of the training process. While some units became sensitive to either iEEG amplitude or phase, others showed bimodal behavior with significant sensitivity to both features. Pruning of highly sensitive units resulted in a steep drop of decoding accuracy not observed for pruning of less sensitive units, highlighting the functional relevance of the amplitude- and phase-specialized populations. Significance. We anticipate that emergent functional specialization as uncovered here will become a key concept in research towards interpretable deep learning for neuroscience and BCI applications
- Location
-
Deutsche Nationalbibliothek Frankfurt am Main
- Extent
-
Online-Ressource
- Language
-
Englisch
- Notes
-
Journal of neural engineering. - 19, 3 (2022) , 036006, ISSN: 1741-2552
- Event
-
Veröffentlichung
- (where)
-
Freiburg
- (who)
-
Universität
- (when)
-
2022
- Creator
-
Hammer, Jiři
Schirrmeister, Robin Tibor
Hartmann, Kay Gregor
Marusic, Petr
Schulze-Bonhage, Andreas
Ball, Tonio
- DOI
-
10.1088/1741-2552/ac6770
- URN
-
urn:nbn:de:bsz:25-freidok-2269851
- Rights
-
Open Access; Der Zugriff auf das Objekt ist unbeschränkt möglich.
- Last update
-
2025-03-25T13:53:06+0100
Data provider
Deutsche Nationalbibliothek. If you have any questions about the object, please contact the data provider.
Associated
- Hammer, Jiři
- Schirrmeister, Robin Tibor
- Hartmann, Kay Gregor
- Marusic, Petr
- Schulze-Bonhage, Andreas
- Ball, Tonio
- Universität
Time of origin
- 2022
Other Objects (12)
Harmonic analysis of deep convolutional neural networks
Deep Convolutional Neural Networks for Biomedical Image Analysis
Theory of Deep Convolutional Neural Networks and Distributed Learning
Automatic electroencephalogram artifact removal using deep convolutional neural networks
Reliable Company Logo Detection with Deep Convolutional Neural Networks
Learning cartographic building generalization with deep convolutional neural networks
Deep Learning Based Classification of Pedestrian Vulnerability Trained on Synthetic Datasets
Deep Learning of Arrhythmia Analysis Based on Convolutional Neural Network
Deep convolutional recurrent neural network for rare acoustic event detection
Functional specialization in the brain : the case of numbers
Learning attribute representations with deep convolutional neural networks for word spotting
Multiple trajectory prediction with deep temporal and spatial convolutional neural networks
Harmonic analysis of deep convolutional neural networks
Deep Convolutional Neural Networks for Biomedical Image Analysis
Theory of Deep Convolutional Neural Networks and Distributed Learning
Automatic electroencephalogram artifact removal using deep convolutional neural networks
Reliable Company Logo Detection with Deep Convolutional Neural Networks
Learning cartographic building generalization with deep convolutional neural networks
Deep Learning Based Classification of Pedestrian Vulnerability Trained on Synthetic Datasets
Deep Learning of Arrhythmia Analysis Based on Convolutional Neural Network
Deep convolutional recurrent neural network for rare acoustic event detection
Functional specialization in the brain : the case of numbers
Learning attribute representations with deep convolutional neural networks for word spotting
Multiple trajectory prediction with deep temporal and spatial convolutional neural networks
Harmonic analysis of deep convolutional neural networks
Deep Convolutional Neural Networks for Biomedical Image Analysis
Theory of Deep Convolutional Neural Networks and Distributed Learning
Automatic electroencephalogram artifact removal using deep convolutional neural networks
Reliable Company Logo Detection with Deep Convolutional Neural Networks
Learning cartographic building generalization with deep convolutional neural networks
Deep Learning Based Classification of Pedestrian Vulnerability Trained on Synthetic Datasets
Deep Learning of Arrhythmia Analysis Based on Convolutional Neural Network
Deep convolutional recurrent neural network for rare acoustic event detection
Functional specialization in the brain : the case of numbers
Learning attribute representations with deep convolutional neural networks for word spotting