EEG-based Assessment of Long-Term Vigilance and Lapses of Attention using a User-Centered Frequency-Tagging Approach

Ladouce, S.; Torre Tresols, J. J.; Le Goff, K.; Dehais, F. · 2024 · Crossref

DOI: 10.1101/2024.12.12.628208

archive: archived pipeline: cataloged

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

Summary

This study addresses the challenge of monitoring vigilance and detecting attentional lapses during prolonged, monotonous tasks, such as surveillance or driving, where performance degradation poses significant safety risks. While electroencephalography (EEG) markers like alpha and theta power have been linked to fatigue, they often lack the specificity and temporal stability required for real-time prediction of individual attentional lapses. The authors propose a minimally intrusive Steady-State Visual Evoked Potential (SSVEP) approach, utilizing a low-contrast, 14 Hz frequency-tagging flicker superimposed on a visual task to continuously track attention without disrupting user experience or task performance. The experimental design involved 16 healthy participants performing two 45-minute sessions of the Mackworth Clock Task, a sustained visual attention paradigm where participants monitor a moving indicator for rare "skip" events. In one session, a transparent 14 Hz flicker was overlaid on the screen; in the other, no flicker was present. EEG data were recorded using a 32-channel system and processed using Rhythmic Entrainment Source Separation (RESS) to extract the SSVEP signal-to-noise ratio (SNR). This SSVEP measure was compared against individual alpha peak frequency (IAPF) and frontal theta band power, which served as baseline spectral markers. Subjective measures, including the Karolinska Sleepiness Scale and NASA Task Load Index, were collected to assess the impact of the flicker on user comfort. Results indicated that the SSVEP response was significantly lower in the periods preceding attentional lapses (missed target events) compared to successful detections. In contrast, traditional spectral markers, specifically IAPF and theta band activity, did not reliably distinguish between missed and detected events. Crucially, the presence of the 14 Hz flicker did not alter task performance (accuracy or reaction times) nor did it increase subjective reports of eye strain, fatigue, or sleepiness. The flicker was designed to be imperceptible or minimally intrusive, ensuring that the neural signal reflected genuine attentional states rather than artifacts of visual distraction. The significance of these findings lies in the demonstration that a non-intrusive, continuous SSVEP-based metric can effectively predict attentional lapses in real-time settings. Unlike broad-band spectral power measures, the frequency-tagged SSVEP offers higher specificity for isolating attentional engagement from general fatigue or time-on-task effects. This approach holds promise for the development of passive brain-computer interfaces that can monitor operator vigilance in high-stakes environments, providing a practical tool for enhancing safety in domains such as air traffic control and long-haul driving.

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 qwen3.8-27b-gittensor on 2026-08-23; verification: pending re-verification.

Topics

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

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

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