Document is current

Article History

Received: 2 January 2026

Accepted: 16 March 2026

First Online: 23 May 2026

Declarations

:

: All studies received institutional review board (IRB) approval prior to initiation: Study 1—Controlled Laboratory Experiments (n = 847): Approved by the University of St. Thomas Human Subjects Research Committee (protocol HSR-2021-047, approved March 2021). Study 2—High-Fidelity Flight Simulations (n = 134): Approved by the Woxsen University Ethics Review Board (protocol WU-2021-23, approved April 2021). Study 3—Operational Field Observations (29 airports; 2.3 million interactions): Approved under the same institutional protocols listed above, with additional organizational data-sharing agreements in place for each participating airport and aviation authority. All studies were conducted in accordance with the Declaration of Helsinki (ethical principles for medical research involving human subjects). Human ethics protocols were followed throughout all phases, including: • Participant recruitment via professional organizations (e.g., Air Line Pilots Association) with clearly defined inclusion/exclusion criteria (current certification, minimum experience thresholds, absence of neurological conditions). • De-identified data collection and secure storage in compliance with GDPR requirements for field data. • Equity and fairness checks confirming no demographic moderation of outcomes (gender: ANOVA F(1,845) = 1.2, p  = 0.27; age: r = 0.08, p  = 0.42). • Privacy-protective monitoring design incorporating opt-in participation and appropriate mitigations, as fully detailed in the Methods section of the manuscript. • Exclusion of all vulnerable populations. • Data retention in accordance with IRB guidelines (90 days for raw behavioral logs post-analysis).All human participants including aviation professionals, pilots, air traffic controllers, dispatchers, and operational operators provided written informed consent prior to their participation in any study activity. Consent forms included the following elements, consistent with institutional guidelines and the Declaration of Helsinki: • Study procedures and tasks: A full description of all experimental, simulation, and observational activities participants were asked to perform. • Risks and benefits: Risks were minimal (fatigue from simulation sessions, no greater than those encountered in routine operations); benefits were identified as contribution to aviation safety research. • Voluntary participation: Explicit statement of the right to withdraw at any time without penalty or consequence to professional standing. • Data handling: Explanation of de-identification procedures, secure storage protocols, and limits on data use. • Contact information: Details of the principal investigator and ethics board contacts for questions or concerns. No participant was enrolled without documented consent. Consent procedures were reviewed and approved by both the University of St. Thomas Human Subjects Research Committee and the Woxsen University Ethics Review Board prior to participant contact.

: All participants provided informed consent for the publication of anonymized, aggregated data arising from their participation. Consent forms explicitly addressed the possibility of publication and confirmed that no individual participant data or personally identifiable information would be disclosed. The following safeguards govern all data presented in this manuscript: • All data presented in figures, tables, and text are aggregated across participants and sites, ensuring that no individual can be identified from any reported result. • No raw participant-level records, identifiers, or attributable quotes are included in the manuscript or supplementary materials. • Participating organizations reviewed and approved the use of operational data in aggregated form for research publication purposes, in accordance with their respective data-sharing and confidentiality agreements. • Datasets are not publicly available due to privacy restrictions and confidentiality obligations; however, aggregated data supporting the findings may be obtained from the corresponding author upon reasonable request and subject to institutional data-sharing agreements.

: The authors declare no competing interests.

: Large language models were consulted for literature review organization and initial manuscript language refinement. All scientific content—including study design, data collection protocols, statistical analysis methods, results interpretation, and conclusions—was developed independently by the authors without AI assistance. No AI tools were used for data analysis, feature engineering, model development, or generation of figures and tables. The machine learning classifiers described in this manuscript (Random Forest, Gradient Boosting, Logistic Regression) are analytical methods applied to empirical data, not AI writing tools. All authors have reviewed, edited, and take full responsibility for the accuracy and integrity of the entire manuscript.