Prediction of Reaction Time and Vigilance Variability From Spatio-Spectral Features of Resting-State EEG in a Long Sustained Attention Task

Torkamani-Azar, Mastaneh; Kanik, Sumeyra Demir; Aydin, Serap; Cetin, Mujdat · 2020 · Crossref

DOI: 10.1109/jbhi.2020.2980056

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study addresses the challenge of predicting sustained attention performance and vigilance variability from pre-task, resting-state electroencephalography (EEG) data. While resting-state brain networks are known to reflect intrinsic cognitive states, prior research lacked concise predictors for task-induced vigilance fluctuations using spatio-spectral EEG features. The authors aimed to develop an automated framework to estimate reaction time and vigilance stability before task execution, which is critical for adaptive brain-computer interfaces (BCIs) and monitoring operators in monotonous, high-stakes environments like air traffic control. The researchers recruited ten healthy volunteers who underwent 105-minute sessions of a Sustained Attention to Response Task (SART). Prior to the task, participants completed 2.5-minute eyes-open and eyes-closed resting-state EEG recordings. The study introduced a novel, adaptive Cumulative Vigilance Score (CVS) based on error rates and hit response times, adjusted for individual response styles to objectively quantify vigilance maintenance without subjective self-reports. Preprocessing involved automated artifact removal and feature extraction, resulting in band-power ratios across 14 regions of interest and 12 frequency bands. The authors employed single-layer neural networks with leave-one-subject-out cross-validation to identify significant features, followed by multiple linear regression to model the relationships between resting-state EEG features and performance metrics, including mean CVS, mean response time, and their respective variabilities. The results demonstrated that specific spatio-spectral features from resting-state EEG significantly predicted task performance. Increased gamma (28–48 Hz) and upper beta (24–28 Hz) power ratios in left central and temporal regions predicted slower reactions and greater vigilance inconsistency, attributed to increased activation of the default mode network. Conversely, higher parietal alpha (8–12 Hz) ratios in Brodmann’s areas 18, 19, and 37 during eyes-open states predicted slower responses but more consistent vigilance scores, indicating superior ability to maintain attention. The neural network analysis revealed distinct associations between performance measures and specific beta sub-bands, while the regression models confirmed that small subsets of intrinsic EEG features could reliably predict overall task stability and speed. These findings provide the first evidence that stable, significant predictors of attention variability can be derived from intrinsic EEG power ratios recorded before task onset. The proposed framework allows for the modeling of attention variations during BCI calibration sessions, enabling systems to adjust interface parameters—such as stimulus intensity or repetition rates—based on an operator’s predicted ability to sustain vigilance. This approach offers a non-invasive, objective method for assessing cognitive readiness and fatigue, potentially enhancing the accuracy and safety of human-computer interactions in critical applications.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-nvidia on 2026-08-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• neuro workload indices
• sustained attention vigilance
• drowsiness detection algorithms

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Empirical Findings: physiological data, behavioral performance data