Vigilance decrements in air traffic control as an aviation safety risk: a systematic review of modeling and prediction approaches
Introduction Vigilance decrements in air traffic control (ATC) are a persistent problem for neuroergonomics: human cognitive performance must remain reliable across long operational shifts in an environment where lapses carry serious consequences. Sustained attentional failure is associated with degraded conflict detection, slower response to critical events, and airspace management errors. Impairment is bidirectional: underloaded controllers disengage while overloaded ones lose resolution, and current monitoring systems do not reliably distinguish between these states. Methods This paper reports a systematic review of vigilance modeling and prediction in ATC, covering 34 articles from three major databases (PubMed, Scopus, Web of Science; initial search January 2025, forward citation update through February 2026), selected through pre-specified inclusion criteria following PRISMA 2020 guidelines. Results The review addresses physiological and behavioral predictors, machine learning detection models, and multimodal fusion approaches for vigilance prediction. Task demand, time on position, circadian phase, and automation-induced passivity all shape vigilance in real operations, and none acts independently; existing predictive frameworks have not captured this structure well. Discussion Each approach is evaluated against the constraints of real-time deployment, and the distance between controlled laboratory findings and tools usable on an operational floor remains substantial.