EEG-based detection of the locus of auditory attention with convolutional neural networks
- Katholieke Universiteit Leuven, Belgium
Abstract
In a multi-speaker scenario, the human auditory system is able to attend to one particular speaker of interest and ignore the others. It has been demonstrated that it is possible to use electroencephalography (EEG) signals to infer to which speaker someone is attending by relating the neural activity to the speech signals. However, classifying auditory attention within a short time interval remains the main challenge. We present a convolutional neural network-based approach to extract the locus of auditory attention (left/right) without knowledge of the speech envelopes. Our results show that it is possible to decode the locus of attention within 1 to 2 s, with a median accuracy of around 81%. These results are promising for neuro-steered noise suppression in hearing aids, in particular in scenarios where per-speaker envelopes are unavailable.
Data availability
Code used for training and evaluating the network has been made available at https://github.com/exporl/locus-of-auditory-attention-cnn. The CNN models used to generate the results shown in the paper are also available at that location. The dataset used in this study had been made available earlier at https://zenodo.org/record/3377911.
-
Auditory Attention Detection Dataset KULeuvenhttp://doi.org/10.5281/zenodo.3377911.
Article and author information
Author details
Servaas Vandecappelle
Alexander Bertrand
Funding
KU Leuven Special Research Fund (C14/16/057)
- Tom Francart
KU Leuven Special Research Fund (C24/18/099)
- Alexander Bertrand
Research Foundation Flanders (1.5.123.16N)
- Alexander Bertrand
Research Foundation Flanders (G0A4918N)
- Alexander Bertrand
European Research Council (637424)
- Tom Francart
European Research Council (802895)
- Alexander Bertrand
The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.
Ethics
Human subjects: The experiment was approved by the Ethics Committee Research UZ/KU Leuven (S57102) and every participant signed an informed consent form approved by the same commitee.
Copyright
© 2021, Vandecappelle et al.
This article is distributed under the terms of the Creative Commons Attribution License permitting unrestricted use and redistribution provided that the original author and source are credited.
Metrics
-
- 3,500
- views
-
- 492
- downloads
-
- 50
- citations
Views, downloads and citations are aggregated across all versions of this paper published by eLife.
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
EEG-based detection of the locus of auditory attention with convolutional neural networks
eLife 10:e56481.
https://doi.org/10.7554/eLife.56481
Further reading
-
- Neuroscience
Previous studies on reinforcement learning have identified three prominent phenomena: (1) individuals with anxiety or depression exhibit a reduced learning rate compared to healthy subjects; (2) learning rates may increase or decrease in environments with rapidly changing (i.e. volatile) or stable feedback conditions, a phenomenon termed learning rate adaptation; and (3) reduced learning rate adaptation is associated with several psychiatric disorders. In other words, multiple learning rate parameters are needed to account for behavioral differences across participant populations and volatility contexts in this flexible learning rate (FLR) model. Here, we propose an alternative explanation, suggesting that behavioral variation across participant populations and volatile contexts arises from the use of mixed decision strategies. To test this hypothesis, we constructed a mixture-of-strategies (MOS) model and used it to analyze the behaviors of 54 healthy controls and 32 patients with anxiety and depression in volatile reversal learning tasks. Compared to the FLR model, the MOS model can reproduce the three classic phenomena by using a single set of strategy preference parameters without introducing any learning rate differences. In addition, the MOS model can successfully account for several novel behavioral patterns that cannot be explained by the FLR model. Preferences for different strategies also predict individual variations in symptom severity. These findings underscore the importance of considering mixed strategy use in human learning and decision-making and suggest atypical strategy preference as a potential mechanism for learning deficits in psychiatric disorders.
-
- Neuroscience
In birds and insects, the female uptakes sperm for a specific duration post-copulation known as the ejaculate holding period (EHP) before expelling unused sperm and the mating plug through sperm ejection. In this study, we found that Drosophila melanogaster females shortens the EHP when incubated with males or mated females shortly after the first mating. This phenomenon, which we termed male-induced EHP shortening (MIES), requires Or47b+ olfactory and ppk23+ gustatory neurons, activated by 2-methyltetracosane and 7-tricosene, respectively. These odorants raise cAMP levels in pC1 neurons, responsible for processing male courtship cues and regulating female mating receptivity. Elevated cAMP levels in pC1 neurons reduce EHP and reinstate their responsiveness to male courtship cues, promoting re-mating with faster sperm ejection. This study established MIES as a genetically tractable model of sexual plasticity with a conserved neural mechanism.
