AUTHOR=Feng Jingwen , Hu Bo , Sun Jingting , Zhang Junpeng , Wang Wen , Cui Guangbin 

TITLE=Identifying Fragmented Reading and Evaluating Its Influence on Cognition Based on Single Trial Electroencephalogram

JOURNAL=Frontiers in Human Neuroscience

VOLUME=Volume 15 - 2021

YEAR=2021

URL=https://www.frontiersin.org/journals/human-neuroscience/articles/10.3389/fnhum.2021.753735

DOI=10.3389/fnhum.2021.753735

ISSN=1662-5161

ABSTRACT=Background: Daily use of social media will nurture a fragmented reading (FR) habit. However, little is known whether FR affect cognition and what are the underlying EEG alterations it may lead to.
Purpose: This study aims to identify whether individuals have FR habit or not based on the single trial EEG spectral features using machine learning (ML), and find out the potential cognitive impairment induced by FR.
Methods: Subjects were recruited through questionnaire and divided into FR and noFR groups according to the time they spent on FR per day. 64-channel EEG were acquired in Continuous Performance Task (CPT) and segmented into 0.5-1.5s post-stimulus epochs under cue and background conditions. The sample sizes were as follows: FR in cue condition, 692 trials; noFR in cue condition, 688 trials; FR in background condition, 561 trials; noFR in background condition, 585 trials. For these single trials, the relative power (RP) of six frequency bands (delta (1-3Hz), theta (4-7Hz), alpha (8-13Hz), beta1 (14-20Hz), beta2 (21-29Hz), lower gamma (30-40Hz)) were extracted as features. After feature selection, the most important feature sets were fed into three ML models (SVM, KNN, Naive Bayes) to perform identification of FR. RP of six frequency bands were also used as feature sets, respectively, to conduct classification tasks.
Results: The classification accuracy reached up to 96.52% in SVM model under cue condition. Specifically, among six frequency bands, most important features were found in alpha and gamma bands. Gamma achieved the highest classification accuracy (86.69% for cue, 86.45% for background). In both conditions, alpha RP in central sites of FR was stronger than noFR (p<0.001). Gamma RP in frontal site of FR was weaker than noFR in background condition (p<0.001), alpha RP in parieto-occipital sites of FR was stronger than noFR in cue condition (p<0.001).
Conclusion: FR can be identified based on single trial EEG evoked by CPT using ML, and the RP of alpha and gamma may reflect the impairment on attention and working memory by FR. FR might lead to cognitive impairment and worth further exploration.