An Approach to Prediction of Mental Resilience in Fighter Pilot

Selection

Krešimir Ćosić

, Siniša Popović

, Marko Šarlija

, Ivan Kesedžić

and Mate Gambiraža

University of Zagreb Faculty of Electrical Engineering and Computing, Unska 3, Zagreb, Croatia

Keywords: Prediction, Mental Resilience, Stress Resilience, Cognitive Resilience, Fatigue Resilience, Fighter Pilot,

Selection, Monitoring, Physiological Features, fNIRS Features, Oculometric Features.

Abstract: It is well known that human factors are associated with a substantial number of military aviation accidents.

To avoid such catastrophic events, continuous enhancement of pilot selection processes using state-of-the-art

scientific research results is needed since traditionally used self-report paper-and-pencil personality measures

are highly prone to self-report bias. Based on our multidisciplinary research on stress, cognitive and fatigue

resilience, we propose an approach for prediction of mental resilience which is based on: stress resilience

features, like startle reactivity, respiratory sinus arrhythmia, cardiac allostasis; cognitive resilience features

related to cognitive workload/overload, situational awareness, cognitive appraisal etc.; and fatigue resilience

features, like dynamics of saccades, eye blink rates etc. These clusters of features mainly reflect interactions

among all human brain regions, and their multimodal fusion may enhance assessment of pilots’ mental

resilience and combat readiness.

1 INTRODUCTION

It is well known that human factors are associated

with a substantial number of military aviation

accidents (de Hoyos, 2019). To avoid such

catastrophic events, continuous enhancement of pilot

selection processes using state-of-the-art scientific

research results is needed since traditionally used

self-report paper-and-pencil personality measures are

highly prone to self-report bias (Birkeland et al.,

2006). More objective and reliable selection

processes can also minimize drop-out rates among

pilot candidates in later and more expensive stages of

training process. Therefore, selection procedures for

military pilots deserve more attention as new

scientific results become available (Broach et al.,

2019). Unfortunately, scientific literature is

remarkably scarce in addressing these crucial aspects

of selection process, despite declaratively

acknowledging the importance of stress resilience,

stress tolerance and stress coping skills for future

https://orcid.org/0000-0002-6356-743X

https://orcid.org/0000-0001-6026-9261

https://orcid.org/0000-0002-7130-9529

https://orcid.org/0000-0001-8052-0599

https://orcid.org/0000-0002-7669-3929

fighter pilots. Empirical support regarding the role of

personality in pilot performance is lacking, and

further improvements in the pilot selection

procedures are expected (Carretta, 2000). Despite the

relatively low predictive validity of personality

assessment instruments (Broach et al., 2019), specific

personality characteristics closely related to stress

resilience were shown to play a significant role in

mission success in military or other dangerous

occupations (Wood et al., 2015).

The features concerning stress and cognitive

resilience, which may have long-term impact on

future pilot’s task performance are neglected in the

pilot’s selection process (Grassmann et al., 2017).

Furthermore, some research recognizes individual

differences in stress and cognitive resilience features

as an important factor that has short-term and

potentially long-term effects on individual task

performance (Staal et al., 2008; Šarlija et al., 2021;

Vine et al., 2015). Therefore, the value of

objectivized assessment of stress and cognitive

Ä ˛Eosi

c, K., Popovi

c, S., Šarlija, M., Kesedži

c, I. and Gambiraža, M.

An Approach to Prediction of Mental Resilience in Fighter Pilot Selection.

DOI: 10.5220/0011963800003622

In Proceedings of the 1st International Conference on Cognitive Aircraft Systems (ICCAS 2022), pages 83-87

ISBN: 978-989-758-657-6

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

resilience in the early phases of pilot selection is not

yet sufficiently recognized. In our recent project and

articles, we have addressed the challenging problem

of stress resilience prediction (Ćosić et al., 2019a;

Ćosić et al., 2019b; Ćosić et al., 2019c; Šarlija et al.,

2021), as well as cognitive resilience (Kesedžić et al.,

2020). Mental resilience in this paper considers stress

and cognitive resilience, as well as influence of

fatigue on successful operations in military and

civilian stressful and unpredictable environments.

Therefore, prediction of mental resilience in the

context of military pilot selection deserves particular

attention. Based on our multidisciplinary research on

stress, cognitive and fatigue resilience, we propose an

approach for prediction of mental resilience which is

based on: stress resilience features, like startle

reactivity, respiratory sinus arrhythmia, cardiac

allostasis; cognitive resilience features related to

cognitive workload/overload, situational awareness,

cognitive appraisal etc.; and fatigue resilience

features, like dynamics of saccades, eye blink rates

etc. These clusters of features mainly reflect

interactions among all human brain regions, and their

multimodal fusion may enhance assessment of pilots’

mental resilience and combat readiness.

2 PREDICTION OF STRESS

RESILIENCE BY

PHYSIOLOGICAL FEATURES

Stress resilience represents a complex biological,

cognitive, emotional, and behavioural phenomenon,

which can be broadly defined, in the context of a

human psychological trait, as the ability of positive

adjustment to adverse events (Russo et al. 2012,

Ćosić et al., 2019a; Ćosić et al., 2019b). Besides

being a protective factor against the development of

stress-related psychopathology (Walker et al., 2017),

resilience is also defined in a task-related context as

the ability of maintaining normal psychological and

physical functioning, when exposed to extraordinary

levels of stress (Russo et al., 2012; Šarlija et al.,

2021). The latter is crucial in the context of highly

demanding, safety-critical and stressful professions,

such as military fighter jet pilots. In addition to the

well-known psychometric self-report tools, stress

resilience assessment can be objectivized by means

of: (1) various metrics that might allow a deep and

accurate insight into the biological factors

contributing to the candidates’ stress resilience (e.g.

brain imaging, fear conditioning, gene analysis, etc.);

and (2) various features based on the analysis of

objectively measurable responses of the peripheral

physiological signals. Due to the cost- and time-

ineffective nature of the first group of metrics, the

latter has been the focus of our most recent research

(Ćosić et al., 2019a; Ćosić et al., 2019b; Šarlija et al.,

2021).

Prediction of stress resilience based on

physiological features includes the analysis of

peripheral physiological signals, such as

electrocardiography, electrodermal activity,

respiration, and orbicularis oculi electromyography

of eyeblink intensity. Experimental protocols for

elicitation, acquisition and analysis of these signals

are related to: (a) resting autonomic functioning; (b)

the startle reflex; and (c) psychobiological allostasis

(Šarlija et al., 2021). Some of the most prominent

specific physiological features for objectivization of

stress resilience assessment, which have confirmed

discriminative power between an a-priori more

resilient and an a-priori less resilient group of

participants, include: respiratory sinus arrhythmia,

which measures heart rate variability in phase with

inhalation and exhalation; startle reactivity, which

measures the strength of reflexive defensive

responding to an aversive unconditioned stimulus,

i.e., abrupt, loud noise; cardiac allostasis, which

measures adaptive reaction to a stressful event,

involving a vigorous cardiac response to stress

coupled with a significant cardiac recovery in the

aftermath (Ćosić et al., 2019a; Ćosić et al., 2019b).

Our recent work proposes a more comprehensive

set of stress resilience features and application of

machine learning models for prediction of task

performance under stress (Šarlija et al., 2021). The

proposed approach could be used in prediction of

performance envelope limits in realistic stressful

occupational settings, such as the military flight

simulator.

3 PREDICTION OF COGNITIVE

RESILIENCE BY FNIRS

FEATURES

Cognitive resilience is the ability to overcome

negative effects of stress on cognitive functioning

(LaRosa, 2017) and depends on individuals’

cognitive appraisal, cognitive skills, as well as neural

interconnectivity, particularly between limbic and

prefrontal brain structures (Arnsten, 2009). Various

mental states are known to impair cognitive

performance and can jeopardize flight safety, such as

cognitive fatigue (Dehais et al., 2018), cognitive

ICCAS 2022 - International Conference on Cognitive Aircraft Systems

decline and overload. An important factor in pilot

selection process is the evaluation of applicants’

cognitive abilities, mostly estimated using a

computerized test battery, which have high validity,

and low cost (Broach et al., 2019). However, new

methods in cognitive load estimation using objective

neurophysiological measures can be added to these

procedures.

Functional near-infrared spectroscopy (fNIRS)

signals have shown correlations between prefrontal

cortex activation and performance on cognitive tasks,

and can be used in prediction of pilots’ mental states

in mitigation of human error (Verdière et al., 2018),

as well as in cognitive load classification (Kesedžić et

al., 2020). The continuous cognitive load estimation

is of special interest in safety-critical professions,

since the excessive or insufficient cognitive load is

associated with decreased task efficiency (Derosière

et al., 2013). Therefore, the prediction of cognitive

resilience is of particular importance in fighter pilot

selection processes since impairments in cognitive

resilience affect pilot’s performance in stressful

conditions. Consequently, there is a growing interest

in development and implementation of various tools

and means for monitoring and prediction of pilot

cognitive performance (Dehais et al., 2018).

Specific cognitive tasks and corresponding

neurophysiological measures which are already used

in aviation research (Verdière et al. 2018; Dehais et

al., 2018), allow the prediction of cognitive resilience.

Experimental measurements concerning the

prediction of cognitive resilience may include generic

cognitive tests, like working memory tests, arithmetic

tests, n-back tests, and a variety of performance tests

on flight simulators in the later phases of pilot

training. Using comprehensive correlation analysis

and machine learning on multimodal stimuli and

corresponding fNIRS features datasets, assessment

and prediction of pilots’ limits concerning cognitive

load/overload, task performance, attention and

fatigue, as well as lack of situational awareness can

be achieved. fNIRS features include classical

oxygenation features, like signal average, peak,

variance, skewness, kurtosis, area under the curve,

slope, etc., extracted from different prefrontal cortex

regions. These features were already used in research

of cognitive skills such as working memory (Fishburn

et al., 2014), fatigue (Skau et al., 2019), cognitive

flexibility (Kalia et al., 2018), mental workload

(Aghajani et al., 2017), as well as cognitive load

(Kesedžić et al., 2020).

4 PREDICTION OF FATIGUE

RESILIENCE BY

OCULOMETRIC FEATURES

Most accidents in aviation are caused by human error,

which is a result of impaired mental performance. In

terms of the air force and selection of fighter pilots,

oculometric features have shown to be useful for

detecting fatigue or high-workload conditions, as well

as, for investigating motion sickness, hypoxia, and

expertise. Fatigue increases the risk of impaired

performance causing accidents in fighter pilots, and it

is of high importance to select resilient individuals to

prevent possible catastrophic outcomes. Human

performance can be predicted using different

oculometric features related to fixations, saccadic

movements, pupillary response, and eye blinking

(Martinez-Marquez et al., 2021).

Fixation-related features are associated with

visual processing and fatigue. Feature such as fixation

duration is a predictor of fatigue development, and

longer fixation duration indicates increased fatigue

(Zargari Marandi et al., 2018). Saccades-related

features such as saccade velocity, saccadic length,

saccade rate, and the number of saccades are

associated with mental workload, lethargy, and

fatigue, and can be used for prediction of fatigue

resilience. Furthermore, pupil size features are

influenced by emotions, muscular fatigue, as well as

cognitive workload (Gambiraža et al., 2021). The

most common blink-related feature is blink rate,

which is associated with stress, fatigue, task demands,

attention, tension etc.

Eye tracking could also predict pilot’s overall

mental resilience based on features of gaze pattern

dynamics, which can be computed in response to a

variety of visual tasks, like antisaccade, free-viewing,

smooth pursuit tasks, etc. Evaluating pilot’s

oculometric performance on such tasks would

separate those candidates who develop fatigue sooner

from those who develop it later, i.e., from those who

are more fatigue resilient. We have also developed

and optimized general stimulation paradigms for

elicitation of facial and eye gaze features for overall

mental resilience prediction (Ćosić et al., 2019b).

5 CONCLUSION

Prediction of future fighter pilots’ stress, cognitive

and fatigue resilience proposed in this paper is

particularly important due to self-report bias of

standardized psychological instruments in selection

An Approach to Prediction of Mental Resilience in Fighter Pilot Selection

process for attractive high-value jobs (Galić et al.,

2012). These multimodal neurophysiological

measurements and analyses have been successfully

applied in selection processes for air traffic

controllers in Croatia (Ćosić et al., 2019b) and we do

believe that similar approach can be successfully

applied in future fighter pilot selection processes. The

2020 report of the National Commission for Military

Aviation Safety to the U.S. President and Congress

(NCMAS, 2020 Dec 1), as well as Croatian military

aviation accidents (Croatia Week, 2014 Aug 6;

Defense Brief, 2020 Jan 27, 2020 May 7;

FlightGlobal, 2010 Sep 24; Reuters 2007 Jul 9), stress

importance of human factors in military aviation.

Therefore, the main objective of this paper is to

stimulate discussions within broader international

military research community to enhance future pilot

monitoring processes across their professional

lifecycle based on innovative neuro-psycho-

physiological assessments, particularly here in

Croatia regarding upcoming Rafale F3-R jet fighters.

ACKNOWLEDGEMENTS

We acknowledge the support of the state Agency

Alan to present this paper at the ICCAS 2022

conference.

REFERENCES

Aghajani, H., Garbey, M., & Omurtag, A. (2017).

Measuring mental workload with EEG+ fNIRS.

Frontiers in Human Neuroscience, 11, 359.

Arnsten, A. F. (2009). Stress signalling pathways that

impair prefrontal cortex structure and function. Nature

Reviews Neuroscience, 10(6), 410-422.

Birkeland, S. A., Manson, T. M., Kisamore, J. L., Brannick,

M. T. & Smith, M. A. (2006). A meta ‐ analytic

investigation of job applicant faking on personality

measures. International Journal of Selection and

Assessment, 14(4), 317-335.

Broach, D., Schroeder, D., & Gildea, K. (2019). Best

practices in pilot selection (No. DOT/FAA/AM-19/06).

United States. Department of Transportation. Federal

Aviation Administration. Office of Aviation. Civil

Aerospace Medical Institute.

Carretta, T. R. (2000). US Air Force pilot selection and

training methods. Aviation, Space, and Environmental

Medicine, 71(9), 950-956.

Ćosić, K., Šarlija, M., Ivkovic, V., Zhang, Q., Strangman,

G., & Popović, S. (2019a). Stress resilience assessment

based on physiological features in selection of air traffic

controllers. IEEE Access, 7, 41989-42005.

Ćosić, K., Popović, S., Šarlija, M., Mijić, I., Kokot, M.,

Kesedžić, I., & Zhang, Q. (2019b). New tools and

methods in selection of air traffic controllers based on

multimodal psychophysiological measurements. IEEE

Access, 7, 174873-174888.

Ćosić, K., Popović, S., Šarlija, M., Mijić, I., Kokot, M.,

Kesedžić, I. (2019c) Multimodal physiological, voice

acoustic, eye gaze and brain imaging features of stress

resilience. NATO Report of the Project

NATO.MD.SFPP 984829 “Multidisciplinary Metrics

for Soldier Resilience Prediction and Training”.

Croatia Week. (2014 Aug 6). Croatian MIG-21 fighter pilot

hailed a hero after ejecting from crashing jet.

https://www.croatiaweek.com/croatian-mig-21-fighter-

pilot-hailed-a-hero-after-ejecting-from-crashing-jet/

de Hoyos, J. U. (2019). Human Factor in Military Aviation

Accidents: A Resume of 31 USAF accidents extracted

from Accident Investigation Board (AIB) public

reports. Independently published.

Defense Brief. (2020 Jan 27). Croatian Air Force Kiowa

helicopter crashes into Adriatic Sea. https://

defbrief.com/2020/01/27/croatian-air-force-kiowa-heli

copter-crashes-into-adriatic-sea/

Defense Brief. (2020 May 7). Two die in crash of Croatian

Air Force Zlin training aircraft. https://defbrief.com/

2020/05/07/two-die-in-crash-of-croatian-air-force-zlin

-training-aircraft/

Dehais, F., Dupres, A., Di Flumeri, G., Verdiere, K.,

Borghini, G., Babiloni, F., & Roy, R. (2018, October).

Monitoring pilot’s cognitive fatigue with engagement

features in simulated and actual flight conditions using

an hybrid fNIRS-EEG passive BCI. In 2018 IEEE

International Conference on Systems, Man, and

Cybernetics (SMC) (pp. 544-549). IEEE.

Derosière, G., Mandrick, K., Dray, G., Ward, T. E., &

Perrey, S. (2013). NIRS-measured prefrontal cortex

activity in neuroergonomics: strengths and weaknesses.

Frontiers in Human Neuroscience, 7, 583.

Fishburn, F. A., Norr, M. E., Medvedev, A. V., & Vaidya,

C. J. (2014). Sensitivity of fNIRS to cognitive state and

load. Frontiers in Human Neuroscience, 8, 76.

FlightGlobal. (2010 Sep 24). Croatian MiG-21 pilots

escape mid-air collision. https://www.flightglobal.com/

croatian-mig-21-pilots-escape-mid-air-collision/9603

9.article

Galić, Z., Jerneić, Ž., & Kovačić, M. P. (2012). Do

applicants fake their personality questionnaire

responses and how successful are their attempts? A case

of military pilot cadet selection. International Journal

of Selection and Assessment, 20(2), 229-241.

Gambiraža, M., Kesedžić, I., Šarlija, M., Popović, S., &

Ćosić, K. Classification of cognitive load based on

oculometric features. In 2021 44th International

Convention on Information, Communication and

Electronic Technology (MIPRO) (pp. 377-382). IEEE.

Grassmann, M., Vlemincx, E., von Leupoldt, A., & Van den

Bergh, O. (2017). Individual differences in

cardiorespiratory measures of mental workload: An

investigation of negative affectivity and cognitive

ICCAS 2022 - International Conference on Cognitive Aircraft Systems

avoidant coping in pilot candidates. Applied

Ergonomics, 59, 274-282.

Kalia, V., Vishwanath, K., Knauft, K., Vellen, B. V. D.,

Luebbe, A., & Williams, A. (2018). Acute stress

attenuates cognitive flexibility in males only: an fNIRS

examination. Frontiers in Psychology, 9, 2084.

Kesedžić, I., Šarlija, M., Božek, J., Popović, S., & Ćosić, K.

(2020). Classification of cognitive load based on

neurophysiological features from functional near-

infrared spectroscopy and electrocardiography signals

on n-back task. IEEE Sensors Journal, 21(13), 14131-

14140.

LaRosa K. (2017) Cognitive resilience. In: Kreutzer J.,

DeLuca J., Caplan B. (eds) Encyclopedia of clinical

neuropsychology. Springer, Cham.

Martinez-Marquez, D., Pingali, S., Panuwatwanich, K.,

Stewart, R. A., & Mohamed, S. (2021). Application of

eye tracking technology in aviation, maritime, and

construction industries: a systematic review. Sensors,

21(13), 4289.

National Commission for Military Aviation Safety

(NCMAS). (2020 Dec 1). Military aviation losses

FY2013-2020. Report to the President and the Congress

of the United States.

Reuters. (2007 Jul 9). Two killed in Croatian military

chopper crash. https://www.reuters.com/article/idUSL

09167014

Russo, S. J., Murrough, J. W., Han, M. H., Charney, D. S.,

& Nestler, E. J. (2012). Neurobiology of resilience.

Nature Neuroscience, 15(11), 1475-1484.

Skau, S., Bunketorp-Käll, L., Kuhn, H. G., & Johansson, B.

(2019). Mental fatigue and functional near-infrared

spectroscopy (fNIRS)–Based assessment of cognitive

performance after mild traumatic brain injury.

Frontiers in Human Neuroscience, 13, 145.

Staal, M. A., Bolton, A. E., Yaroush, R. A., & Bourne Jr, L.

E. (2008). Cognitive performance and resilience to

stress. In B. J. Lukey & V. Tepe (Eds.). Biobehavioral

resilience to stress (pp. 259-99). Boca Ratton, FL: CRC

Press.

Šarlija, M., Popović, S., Jagodić, M., Jovanovic, T.,

Ivkovic, V., Zhang, Q., & Ćosić, K. (2021). Prediction

of task performance from physiological features of

stress resilience. IEEE Journal of Biomedical and

Health Informatics, 25(6), 2150-2161.

Verdière, K. J., Roy, R. N., & Dehais, F. (2018). Detecting

pilot's engagement using fNIRS connectivity features in

an automated vs. manual landing scenario. Frontiers in

Human Neuroscience, 12, 6.

Vine, S. J., Uiga, L., Lavric, A., Moore, L. J., Tsaneva-

Atanasova, K., & Wilson, M. R. (2015). Individual

reactions to stress predict performance during a critical

aviation incident. Anxiety, Stress, & Coping, 28(4),

467-477.

Walker, F. R., Pfingst, K., Carnevali, L., Sgoifo, A., &

Nalivaiko, E. (2017). In the search for integrative

biomarker of resilience to psychological stress.

Neuroscience & Biobehavioral Reviews, 74, 310-320.

Wood, J., Shurlow, C., & Haynes, J. (2015). Objective

versus subjective military pilot selection methods in the

United States of America. School of Aerospace

Medicine Wright-Patterson AFB OH.

Zargari Marandi, R., Madeleine, P., Omland, Ø., Vuillerme,

N., & Samani, A. (2018). Eye movement characteristics

reflected fatigue development in both young and elderly

individuals. Scientific Reports, 8(1), 1-10.

An Approach to Prediction of Mental Resilience in Fighter Pilot Selection