University of Groningen
Detecting Mind-Wandering with Machine Learning Jin, Christina
DOI:
10.33612/diss.171835555
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2021
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Jin, C. (2021). Detecting Mind-Wandering with Machine Learning: Discovering the Neural Correlates of Mind-Wandering Through Generalizable Machine Learning Classifiers with EEG. University of Groningen. https://doi.org/10.33612/diss.171835555
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
1. Mind-wandering describes task-irrelevant thought that is generated without deliberate control and independent of external cues.
2. Mind-wandering is supported by two key mental processes: sensory decoupling and memory retrieval.
3. Power enhancement in the alpha band (8.5-12Hz), especially observed in occipital regions, is the most important EEG marker of mind-wandering.
4. Although often being reported as being related, mind-wandering, low vigilance and performing an easy task can be distinguished when looking at their neural correlates.
5. Neural signatures of mind wandering are not linearly related, as support vector machines perform better than logistical regression models.
6. Convolutional neural networks perform better with raw EEG signals while support vector machines learn better with pre-computed EEG markers.
7. Given propositions 5 and 6, the complexity of EEG data is better handled by advanced machine learning classifiers.
8. Generalization across various contexts (e.g., individuals, tasks, studies) is important for EEG classifiers to be used in daily life or for therapeutic purposes.
9. We could control mind-wandering in daily-life by using
neurofeedback with machine learning models of EEG. However, healthy mind-wandering helps us with memory consolidation and problem-solving.
10. It is important for researchers to discriminate between mind-wandering’s adaptive and maladaptive forms.
11. The dual-tasking framework of mind-wandering might be the closest description of how it fluctuates.
