An Effective Entropy-Assisted Mind-Wandering Detection System Using EEG Signals of MM-SART Database

Chen, Yi-Ta; Lee, Hsing-Hao; Shih, Ching-Yen; Chen, Zih-Ling; Beh, Win-Ken; Yeh, Su‐Ling; Wu, An-Yeu

doi:10.1109/jbhi.2022.3187346

Cited by 12 publications

(14 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Permutation entropy is an information complexity measure for time-series (Bandt & Pompe, 2002) which indexes the degree of conscious awareness in controls compared to anesthetized and minimally conscious state patients (Thul et al, 2016). Relatedly, Chen et al (2020) used several different classifiers (e.g., SVM, random forest, naive Bayes, and k-nearest neighbors) with standard spectral measures as well as spectral entropy measures and found that the random forest classifier fared best with entropy-based features. Increased measures of complexity, including permutation entropy, have been suggested to be necessary for a specific representation to be selected for conscious processing (Sitt et al, 2014).…”

Section: Discussionmentioning

confidence: 99%

“…Altogether, findings from electrophysiological studies point towards a reliable EEG signature for TUTs which can be readily exploited with machine learning techniques. Accordingly, recent research demonstrates the possibility of predicting the occurrence of TUTs based on EEG measures (Chen et al, 2020; Dhindsa et al, 2019; Dong et al, 2021; Jin et al, 2019; Polychroni et al, 2022). Different classification approaches have been utilized on varied features of EEG and despite some variations, certain features appear most characteristic of TUTs, e.g., P3 and alpha (Dong et al, 2021; Groot et al, 2021; Polychroni et al, 2022), corroborating electrophysiological studies.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Distinct Electrophysiological Signatures of Intentional and Unintentional Mind-Wandering Revealed by Low-Frequency EEG Markers

Martel

Bruno

Robertson

et al. 2023

Preprint

View full text Add to dashboard Cite

Mind-wandering is typically characterized by the common experience wherein attention veers off into thoughts unrelated to the task at hand. Recent research highlights the intentionality dimension of mind-wandering as a key predictor of adverse functional outcomes with intentional and unintentional task-unrelated thought (TUT) differentially linked to neural, behavioral, clinical, and functional correlates. We here aimed to elucidate the electrophysiological underpinnings of intentional and unintentional TUT by systematically examining the individual and collective discriminative power of a large set of EEG markers to distinguish between attentional states. Univariate and multivariate analyses were conducted on 54 predefined markers belonging to four conceptual families: ERP, spectral, information theory and connectivity measures, extracted from scalp EEG recordings prior to multidimensional reports of ongoing thought from participants performing a sustained attention task. We report here that on-task, intentional and unintentional TUT exhibit distinct electrophysiological signatures in the low frequency range. More specifically, increased features of the theta frequency range were found to be most discriminative between on-task and off-task states, while features within the alpha band were characteristic of intentional TUT when compared to unintentional TUT. This result is theoretically well aligned with contemporary accounts describing alpha activity as an index of internally oriented attention and a potential mechanism to shield internal processes from sensory input. Our study verifies the validity of the intentionality dimension of mind-wandering and represents a step forward towards real-time detection and mitigation of maladaptive mind-wandering.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Distinct Electrophysiological Signatures of Intentional and Unintentional Mind-Wandering Revealed by Low-Frequency EEG Markers

Martel

Bruno

Robertson

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Permutation entropy is an information complexity measure for time-series (94), which indexes the degree of conscious awareness in controls compared to anesthetized and minimally conscious state patients (71). Relatedly, Chen et al (57) used several different classifiers (e.g., SVM, random forest, naive Bayes, and k-nearest neighbors) with standard spectral measures as well as spectral entropy measures and found that the random forest classifier performed best with entropy-based features. Increased measures of complexity, including permutation entropy, have been suggested to be necessary for a specific representation to be selected for conscious processing (59).…”

Section: Discussionmentioning

confidence: 99%

“…Electrophysiological study findings collectively suggest the existence of a reliable EEG signature for TUT that can be effectively utilized using machine learning techniques. Recent studies have demonstrated the potential to predict TUT occurrence using EEG measures (33,43,50,57).…”

Section: Introductionmentioning

confidence: 99%

Distinct Electrophysiological Signatures of Intentional and Unintentional Mind-Wandering Revealed by Low-Frequency EEG Markers

Bruno

Dockree

Robertson³

et al. 2023

Preprint

View full text Add to dashboard Cite

Mind-wandering, a widespread mental phenomenon in which attention shifts from an ongoing task, varies in intentionality. Recent research highlights the importance of differentiating between intentional and unintentional mind-wandering, as the latter is uniquely linked to adverse outcomes such as psychopathologies. In this study, we explored the electrophysiological underpinnings of these distinct forms of mindwandering using a robust feature extraction tool adapted from research on neural signatures of consciousness. We conducted univariate and multivariate pattern analyses on 54 EEG markers obtained from recordings immediately before participants provided multidimensional reports of their thoughts during a sustained attention task. Our findings revealed distinct electrophysiological signatures for on-task, intentional, and unintentional mind-wandering states, particularly within the low-frequency spectrum. Specifically, normalized theta power demonstrated the highest discriminative power for discerning on- and off-task states, while alpha band features and theta permutation entropy were uniquely associated with intentional versus unintentional mind-wandering. These results challenge the prevailing notion that increased alpha band activity is a generic marker of mind-wandering and suggest that unique brain activity patterns underlie the various forms of mind-wandering. Our study lays the groundwork for developing reliable, real-time detection systems for identifying mind-wandering using EEG machine learning models in both clinical and practical settings.

show abstract

“…Many studies have tried to detect the mind wandering state, however, they were conducted in laboratory settings with well-controlled stimuli ( Jin et al, 2019 ; Dong et al, 2021 ; Groot et al, 2021 ; Chen et al, 2022 ). Particularly, Chen et al (2022) designed a multi-modal sustained attention to response task (MM-SART), in which participants were instructed to press a key when non-target stimuli appeared and to refrain from doing so when target stimuli appeared. The mental states of the participants were measured via thought probes administered at the end of each experimental block.…”

Section: Introductionmentioning

confidence: 99%

Mind wandering state detection during video-based learning via EEG

Tang

Liang

Zheng

2023

Front. Hum. Neurosci.

View full text Add to dashboard Cite

The aim of this study is to explore the potential of technology for detecting mind wandering, particularly during video-based distance learning, with the ultimate benefit of improving learning outcomes. To overcome the challenges of previous mind wandering research in ecological validity, sample balance, and dataset size, this study utilized practical electroencephalography (EEG) recording hardware and designed a paradigm consisting of viewing short-duration video lectures under a focused learning condition and a future planning condition. Participants estimated statistics of their attentional state at the end of each video, and we combined this rating scale feedback with self-caught key press responses during video watching to obtain binary labels for classifier training. EEG was recorded using an 8-channel system, and spatial covariance features processed by Riemannian geometry were employed. The results demonstrate that a radial basis function kernel support vector machine classifier, using Riemannian-processed covariance features from delta, theta, alpha, and beta bands, can detect mind wandering with a mean area under the receiver operating characteristic curve (AUC) of 0.876 for within-participant classification and AUC of 0.703 for cross-lecture classification. Furthermore, our results suggest that a short duration of training data is sufficient to train a classifier for online decoding, as cross-lecture classification remained at an average AUC of 0.689 when using 70% of the training set (about 9 min). The findings highlight the potential for practical EEG hardware in detecting mind wandering with high accuracy, which has potential application to improving learning outcomes during video-based distance learning.

show abstract

An Effective Entropy-Assisted Mind-Wandering Detection System Using EEG Signals of MM-SART Database

Cited by 12 publications

References 40 publications

Distinct Electrophysiological Signatures of Intentional and Unintentional Mind-Wandering Revealed by Low-Frequency EEG Markers

Distinct Electrophysiological Signatures of Intentional and Unintentional Mind-Wandering Revealed by Low-Frequency EEG Markers

Distinct Electrophysiological Signatures of Intentional and Unintentional Mind-Wandering Revealed by Low-Frequency EEG Markers

Mind wandering state detection during video-based learning via EEG

Contact Info

Product

Resources

About