Combining EEG and eye-tracking for cognitive and physiological states monitoring: a systematic review

Rivas-Vidal, Maria; Calvo Cordoba, Alberto; García Cena, Cecilia E.; Farfán, Fernando Daniel · 2026 · Crossref

DOI: 10.3389/fnrgo.2025.1736672

archive: archived pipeline: cataloged

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

Summary

This systematic review addresses the gap in understanding how electroencephalography (EEG) and eye-tracking (ET) jointly characterize perception-related cognitive states, such as mental workload, fatigue, stress, drowsiness, and vigilance. While these non-invasive modalities offer complementary insights into neural and ocular signatures of attention, prior literature typically examined them in isolation or focused on single conditions, limiting the ability to distinguish shared physiological patterns. The study aims to synthesize evidence from concurrent EEG-ET recordings to identify common features and evaluate the efficacy of multimodal classification models for monitoring situational awareness in high-risk environments. Following the PRISMA 2020 guidelines, the authors searched five databases (PubMed, Scopus, Cochrane, IEEE Xplore, and Web of Science) for studies published up to October 2024. From 581 initial records, 47 studies met the inclusion criteria, requiring the co-registration of EEG and ET data to quantify specific cognitive conditions in healthy adult populations. The review employed a descriptive synthesis approach, extracting data on experimental designs, feature extraction, and classification performance. Risk of bias was assessed using the ROBINS-I tool, with most studies rated as having moderate risk, primarily due to insufficient reporting on confounding factors and participant selection. The analysis revealed that theta, alpha, and beta EEG activity, along with ET metrics such as fixation patterns, pupil diameter, blink dynamics, and percentage of eyes closed (PERCLOS), were the most frequently reported features. The findings indicate that fatigue, mental workload, and stress exhibit overlapping physiological signatures, whereas drowsiness and vigilance decrement appear along a shared continuum, with microsleeps displaying distinct profiles. Crucially, classification models integrating both EEG and ET features generally achieved higher accuracy than those relying on a single modality. This multimodal integration helps disambiguate closely related states that are difficult to distinguish using unimodal data alone. The significance of this review lies in its support for the development of standardized multimodal protocols and real-time classification models for cognitive-state monitoring. By demonstrating the superior discriminative power of combined EEG-ET systems, the study highlights the potential to enhance operational performance and error prevention in domains requiring sustained attention, such as military, aerospace, and industrial operations. The authors conclude that while current evidence supports the utility of multimodal monitoring, further research is needed to establish unified frameworks that can reliably detect perception-related conditions in real-world settings.

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
• drowsiness detection algorithms
• workload measurement
• drowsiness
• gaze based attention detection

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
• Methodological Resource: tool software, validation psychometrics