A novel EEG-driven multidimensional modelling of the Driver Performance Envelope for a user-centred human-vehicle interaction

Giorgi, Andrea; Ronca, Vincenzo; Dello Iacono, Francesca; Cecchetti, Marianna; Capotorto, Rossella; Menicocci, Stefano; Rossi, Dario; Aricò, Pietro; Borghini, Gianluca; Biassoni, Federica; Bina, Manuela; Babiloni, Fabio; Di Flumeri, Gianluca · 2026 · Crossref

DOI: 10.54941/ahfe1007860

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 need for continuous, objective assessment of driver capability by proposing a novel multidimensional Driver Performance Envelope (DPE) model. Adapted from the Human Performance Envelope concept used in aviation, the DPE defines the boundary within which a driver can perform safely. The authors argue that current driving performance models rely heavily on behavioral indicators, lacking the neurophysiological sensitivity to detect rapid fluctuations in mental states such as workload, stress, vigilance, and fatigue. To bridge this gap, the research develops an EEG-driven framework that provides time-resolved estimation of a driver’s readiness to operate a vehicle, aiming to support user-centered human-vehicle interactions and enhance road safety. The experimental design involved fifteen participants who completed five simulated driving scenarios designed to elicit varying levels of performance capacity: two high-DPE conditions (clear weather, low traffic), one intermediate-DPE condition (urban driving with time pressure), and two low-DPE conditions (high traffic, poor visibility, and secondary tasks). Continuous EEG data were recorded using a wearable eight-channel system. The signal processing pipeline included artifact removal and the extraction of validated neurometrics for mental effort, vigilance, stress, and fatigue. These metrics were integrated into a composite DPE index using Principal Component Analysis (PCA) to capture shared variance, with Mutual Information (MI) employed to verify the reliability and coherence of the underlying neurometrics. Subjective questionnaires were administered after each scenario to validate the experimental manipulations. The results demonstrated that the EEG-driven DPE index successfully discriminated between driving conditions with different hypothesized performance envelopes. The DPE values were significantly higher in the high-performance conditions compared to intermediate and low-performance scenarios, aligning with the experimental hypotheses. Subjective assessments confirmed that low-DPE tasks induced significantly higher perceived difficulty, stress, and fatigue than high-DPE tasks. Notably, the MI analysis revealed that the neurometrics shared significant information only during active driving tasks, while MI values were significantly lower during resting-state conditions, confirming that the PCA-derived index loses sensitivity when the target phenomenon (driving performance) is absent. One low-DPE condition yielded unexpectedly high DPE values, which the authors attribute to participants finding the high-speed challenge engaging rather than overwhelming. The study concludes that the proposed DPE model is a feasible and robust tool for continuously monitoring driver psychophysiological states. By integrating neurophysiological evidence into human factors modeling, the framework offers a promising method for assessing driver fitness in real-time. This approach has significant implications for developing adaptive automation strategies in next-generation vehicles, where system behavior can be tailored to the driver’s current capability envelope rather than relying solely on contextual data, thereby reducing human error and enhancing overall mobility safety.

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.

• drowsiness detection algorithms
• stress driving
• mental demand
• workload measurement
• cognitive capacity variation
• neuro workload indices

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: validation psychometrics
• Theoretical Contribution: computational model