Near-invisible c-VEP-based passive BCI for mental workload monitoring

Pietro, Cimarosto; Sebastien, Velut; Kalou, Castillos-Cabrera; Tresols Juan, Torre; Raphaëlle N, Roy; Frederic, Dehais · 2026 · Crossref

DOI: 10.1088/1741-2552/ae4ff6

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 adapting visual evoked potential (VEP)-based Brain-Computer Interfaces (BCIs) for passive mental workload monitoring. While VEPs are robust for reactive BCIs, their use in passive BCIs is limited by the distracting nature of flickering stimuli, which can impair visual comfort and divert attention from primary tasks. The authors propose an unobtrusive approach using "Stimuli for Augmented Response" (StAR)—near-invisible, texture-based flickers overlaid on user interface regions of interest—combined with a code-based VEP (c-VEP) pipeline to assess cognitive load without requiring intentional fixation. The research involved 20 healthy participants who completed two experimental sessions: one in a laboratory setting using the Multi-Attribute Task Battery II (MATB-II) and another in a flight simulator. In both environments, participants performed multitasking duties under three workload conditions: Supervision (low load, passive observation), Easy, and Hard. StAR stimuli were presented on task-relevant displays with reduced luminance to enhance comfort. EEG data were recorded using 32-channel gel-based electrodes, and eye-tracking data were collected to monitor gaze and artifacts. The c-VEP pipeline utilized Riemannian geometry-based decoding to detect single-trial responses, computing a "coherence" index that measured the correlation between predicted and reference binary codes. This coherence metric served as a continuous proxy for mental workload, under the hypothesis that higher cognitive demand reduces ERP amplitude and detection reliability. Results demonstrated that the amplitude of visual event-related potentials (ERPs), specifically N200 and P300 components, was significantly reduced under higher workload conditions at the group level. This neural marker provided a reliable basis for workload assessment. Furthermore, the proposed c-VEP pipeline successfully derived coherence indexes that were sensitive to workload-related modulation. The system effectively distinguished between low and high workload states in both the controlled lab environment and the more complex, artifact-prone flight simulator. The study confirmed that the StAR stimuli maintained visual comfort while eliciting robust cortical responses, validating the feasibility of using near-invisible flickers for passive monitoring. The significance of these findings lies in demonstrating that textured, near-invisible flickers combined with c-VEP decoding can effectively monitor cognitive workload in complex operational environments. By mitigating the distraction and discomfort associated with traditional visual flickers, this approach enables the deployment of passive BCIs in real-world scenarios, such as aviation, where continuous, unobtrusive monitoring of pilot mental state is critical for safety and performance optimization. The work bridges the gap between laboratory-validated VEP paradigms and practical, ecologically valid 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
• workload measurement

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