Content uploaded by Gabriela Kloudová
Author content
All content in this area was uploaded by Gabriela Kloudová on Apr 12, 2023
Content may be subject to copyright.
Interservice/Industry Training, Simulation, and Education Conference (I/ITSEC) 2016
2016 Paper No. 16080 Page 1 of 9
Mental Workload of Military Pilots as measured in a Tactical Simulator
Gabriela Kloudova
Dr. Miloslav Stehlik
CASRI – Czech Army Sport Research Institute
CASRI – Czech Army Sport Research Institute
Prague, Czech Republic
Prague, Czech Republic
kloudova@casri.cz
stehlik@casri.cz
ABSTRACT
An overly burdensome mental workload significantly decreases the efficiency of human performance. Therefore, its
realistic replication plays a crucial role in the training of military pilots and helps improve the simulation of
demanding situations and optimize training. Military pilots perform at the extreme limits of physical and mental
human capacity, so it is vital to identify the impact of the mental workload that occurs during highly demanding
flight tasks.
This paper details the research we conducted, as well as results that suggest the best outcomes of those
psychophysiological methods suitable for flight activity in measuring heart rate and electroencephalography, among
other methods used such as eye tracking and biochemistry. This experimental study was held in a tactical flight
simulator designed for fighter aircraft training. The simulator used was a high-level full-flight simulator and is
internationally acknowledged, although it is not equipped with facilities to simulate mental workload and to monitor
pilot resilience. The measurements of physiological data were supported by a psychological questionnaire detecting
subjective perceptions of the mental workload under difficult situations.
Finally, the paper presents results that illustrate the most demanding and difficult situations that appeared during
tactical combat missions and significant results that prove the usability of heart rate monitoring and EEG. We were
also able to distinguish those EEG waves that were most important in identifying the mental workload. This then
enabled us to develop a mental workload index that shows with a high degree of accuracy the most demanding
elements of simulated tactical combat missions. These conclusions contributed to the improvement of training and
were subsequently used under actual flight conditions with helicopter pilots.
ABOUT THE AUTHORS
Gabriela Kloudova is military and sport psychologist specialized in human performance analysis and peak
performance in military aircraft. She is currently working in the research center of the Czech Ministry of Defense as
a Research Psychologist. Her main area of expertise is in work with psychophysiological methods used for
increasing the mental resilience of top athletes, military pilots, and other soldiers performing in extreme conditions.
Miloslav Stehlik is a clinical psychologist and the head of the Department of Psychology in the research center of
the Czech Ministry of Defense. Dr. Stehlik is also currently a Project Manager for a military research program and a
Lecturer at the Czech Military College.
Interservice/Industry Training, Simulation, and Education Conference (I/ITSEC) 2016
2016 Paper No. 16080 Page 2 of 9
Mental workload of military pilots measured in a tactical simulator
Gabriela Kloudova
Miloslav Stehlik
Stehlik Miloslav
cd
Author3 Name
CASRI – Czech Army Sport Research Institute
CASRI – Czech Army Sport Research Institute
Author3 Affiliation
Prague, Czech Republic
Prague, Czech Republic
City, State
kloudova@casri.cz
stehlik@casri.cz
Author3@email.com
INTRODUCTION
Military aviation is cognitively demanding, and measuring pilot mental workload can be an important tool to reduce
the high proportion of aviation accidents (80-85%) that are caused by human error (Thomas & Russo, 2007).
Technological progress in aviation forces pilots to mobilize all of their capacities such as visual, auditory, and
proprioceptive combined with high levels of mental activity. Currently, the top level of training is tactical flight
which requires proficient, adaptive pilots capable of individually solving any problems and making the optimal
decisions regarding consequences. At this point in training, the aim is not to train any specific skill but to build up
mental resilience, producing pilots of strong personality capable of making their own decisions under very short
time constraints - that will also be the most effective. This is especially true during the unexpected occurrence of
new operative demands. For this purpose, we need to make detailed assessments of mental workload. Although
mental workload has been often used for the evaluation of pilot performance, it has usually been based on the pilots’
own subjective reporting (Di Stasi, Antoli, & Canas, 2011). This evaluation is very limited. It is better to combine
such an evaluation with other methods and to use a psychophysiological measurement of mental workload, which
offers a nonintrusive way of continuously collecting data during the complex and dynamic tasks of tactical flight.
Indeed, the noninvasive monitoring of mental workload is a current goal of international organizations worldwide,
especially in critical-safety environments (Friedl, Grate, Proctor, Ness, Lukey, & Kane, 2007; Tracey & Flower,
2014). Understanding the mental workload of humans during flight tasks could be very useful for developing a class
of devices that could alert the pilot or ground control to any low level of internal cognitive resources during flight.
The most frequently used psychophysiological measure to study the effects of mental workload on pilot performance
is heart rate (Jorna, 1993; Kakimoto, Nakamura, & Tarui, 1988; Wilson, 2002). Increased heart rate (HR) has been
observed in various situations encountered by pilots requiring mental work, during both simulated and actual flight
(Hart & Hausser, 1987), and these are considered to reflect workload related to the corresponding flight sequences.
We can see significant changes during highly demanding tasks (Veltman & Gaillard, 1993), but these can be
contaminated by artifacts from muscular activity, which makes the data obtained less reliable. Other
psychophysiological measures used to identify the demands of mental workload included electrodermal activity
(EDA), electroocular activity (EOG), and biochemistry (Wilson, 2002; Boucsein, 2012). Electrodermal activity is
also influenced by muscular activity, while biochemistry does not allow the continuous evaluation of mental
workload (Karthikeyan, Murugappan, & Yaacob, 2013). The most reliable contemporary method to assess mental
workload are electroencephalographic (EEG) recordings, though EEG activity has mainly been recorded in the
laboratory during specific mental tasks and only a few in-flight EEG recordings have been made in a real aircraft
cockpit, given the environmental constraints (Dahlstrom, Nahlinder, Wilson, & Svensson, 2011). Its advantage lies
in the possibility of online recording without interfering with the flight tasks at hand (Borghini, Astolfi, Vecchiato,
Mattia, & Babiloni, 2014). The main studies show that activity in beta (, 13–30 Hz) indicates a cognitive load,
anxiety, and signs of acute stress (Budzynski, Budzynski, Evans, & Abarbanel, 2009; Yi, et al., 2015). Delta band
activity increases during alertness and reaction ground tasks (Dussault, Guezennec, & Jouanin, 2004) and can also
indicate fatigue (Lal & Craig, 2002). Theta (θ, 4–8 Hz) and alpha (, 8-13 Hz) bands tend to intensify during
relaxation and suppress in anxiety (Kubota, Sato, Toichi, Murai, Okada, & Hayashi, 2001; Jacobs & Friedman,
2004; Rauch, Karpul, Derman, & Prinsloo, 2013). A few studies have focused on gamma (γ, 30–70 Hz) band
activities that are probably related to vigilance level, memory, situational awareness, and other important cognitive
tasks (Sokhadze, 2012). Altough the EEG recording is sensitive to mental workload there are certain technical and
methodological difficulties that complicate its use during actual flight.
Our research project was primarily focused on the investigation of the usefulness of psychophysiological methods to
obtain reliable data for a pilot’s mental workload during a simulated tactical mission. After conducting pre-
Interservice/Industry Training, Simulation, and Education Conference (I/ITSEC) 2016
2016 Paper No. 16080 Page 3 of 9
experiments in which we used electrodermal activity, biochemistry, eye tracking, heart rate, heart rate variability,
and EEG, we narrowed down the most relevant methods that can be used in a tactical simulator. Finally, we used
heart rate and EEG, as they appeared to be most suitable for this specific task. This is because both are nonintrusive,
and they produce the greatest information value regarding mental workload.
Subjects
Fifteen experienced military pilots participated in this study, and they trained on the JAS-39 Gripen fighter aircraft.
All subjects were men aged 35-43. They volunteered to participate in the study, had normal vision, and underwent a
full physical examination. All subjects were currently on flight status, indicating recent good health, were
nonsmokers, and were right-hand dominant. None of them took medication for chronic disease or had any history of
cardiac disorders.
Methods
Electroencephalogram: The electroencephalogram signals (EEG)
were monopolarly recorded with a FlexiCap (Deymed) electrode cap,
which holds 19 electrodes according to the international 10/20
system (see Figure 1). It was recorded with a Siesta 802 ambulatory
digital device.
Electrocardiogram: The heart rate was recorded by using a DiANS
PF 8 one-channel electrocardiogram (with two electrodes placed on
the sternum of the subjects), which can be used for capturing heart
rate variability and providing wireless online data collection.
Questionnaire: Post-flight, a questionnaire was used to collect pilots’
subjective ratings on their mental workload and task difficulty
experiences during specific parts of the tactical mission. It also
contained questions about social support from family and friends and
relationships with co-workers and commanders. Pilots could also
comment on their current actual mental and physical health. Another
questionnaire was created for the instructor to evaluate pilot
performance, so all of the data could be compared. In this
questionnaire, the instructor could evaluate every situation that occurred during simulated flight on a scale from 1-5
and add some comments and observations.
Study Protocol
This experimental study was conducted in the Tactical Simulation Centre in Pardubice, Czech Republic (TSC). This
facility is part of the Centre of Aviation Training and is officially used for the tactical training of Czech and other
NATO countries’ aviation personnel. Mainly, it is focused on the JAS-Gripen 39 supersonic aircraft, but it is also
suitable for training on other types of subsonic aircraft, such as the L-159. The simulated missions are only tactical
in scope, including all types of air-to-air operations but without the ability to train on common procedures like take-
offs or landings. The TSC can train up to 8 pilots in one simulated combat mission, 2 ground-controlled interception
(GCI) drivers and other operators, and training instructors.
The pilots tested came to the TSC in the morning - well rested, in good health, and without having consumed any
prohibited substances. In case of increased physical activity, they were able to take a rest before the testing. Once
the pilot received his preflight briefing, he was taken to the examination room, and the electrodes of ECG were
placed on his sternum. We secured the optimal conditions for the measurement of heart rate, which occurs 1.5-2
hours after eating, at a constant temperature 20-22˚, and without any medication (Bayevsky, et al., 2002). When the
DiANS PF 8 device was successfully connected to the electrodes, the measurement started. To gain the resting heart
rate, the pilot was instructed to relax and sit quietly for 5 minutes. In the second phase of the experiment, the EEG
electrodes were placed on the scalp of the subject using a special EEG cap and a conductive gel. Use of the EEG
requires a calibration that consists of checking the right impedance and later a 2-minute resting phase with open eyes
Figure 1. Placement of the electrodes
according to the international 10/20
system
Interservice/Industry Training, Simulation, and Education Conference (I/ITSEC) 2016
2016 Paper No. 16080 Page 4 of 9
and a 2-minute phase of closed eyes. Right after verification of the signal, the pilot could go directly to a simulated
cockpit. When all of the devices and personnel were synchronized, the simulated combat mission would start, and it
usually lasted about 20-30 minutes. In total, the pilots flew 22 missions. Four pilots participated in each of them, so
the total number of missions was 88. The pilots were flying tactical missions, including pre-approved scenarios
mainly focused on air defense. The instructor who led the training selected the most frequently demanding
situations, such as engaging the Electronic Warfare System, Ground Control Intercept, Pump (Pitbull), Pump
(Cheapshot), Crossing Flot, and Threatcall. When the mission ended, the pilots stayed in the cockpit, where the
flight questionnaire that they needed to fill out was shown on the main screen. This took about 10-20 minutes. Then
they went back to the examination room where all of the devices were removed, and the data obtained was saved.
The evaluation questionnaire was filled out by the instructor after the training, based on continuous notes and
recordings.
Results
This paper describes simulator-based results regarding the pilots’ mental workload. The combat missions were
randomly ordered each day for each pilot, so it wasn’t possible to ensure that all of the pilots would fly the same
scenario. We monitored the mental workload of the pilots during tactical combat missions by EEG, ECG, and the
subjective questionnaire. Analysis of the EEG data was made by correlation between demanding situations. Firstly,
it was necessary to eliminate eye movement and muscle activity artifacts, which mostly occurred during the
measurements. Eye movements and blinking occurred throughout the awake phase of the subjects when the
measurements were made. These types of artifacts didn’t significantly interfere with the features counted. However,
it was necessary to pay attention to the segments with excessive muscle activity artifacts that were eliminated off-
line before data averaging by a combination of manual evaluation and a semi-automatic method. This was used to
distinguish between the artifact and non-artifact parts of the EEG recordings, and the 2-minute pre-stimulus baseline
was used in all analyses (see Figure 2).
Figure 2. Parts of the recording devaluated by artifacts (black color)
Later on, a selection of the 20 most significant features was made by Sequential Forward Selection (SFS) which
enables finding a combination of uncorrelated features. The correlated situations were the resting phase and the
demanding situations. This selection produces an easier classification of the data and increases the accuracy of the
conclusions made from the recordings. The most significant feature in this case was the O2 electrode on the gamma
waveband (see Table 1). As we can see from Table 1 if we use relative power in the gamma waveband for channel
02, we can reach a 0.7602 classification accuracy that the mental workload measured with this feature was really
experienced by the pilot. This accuracy has, up till now, been hard to reach when using EEG recordings for the
classification of mental workload while using only one waveband.
Interservice/Industry Training, Simulation, and Education Conference (I/ITSEC) 2016
2016 Paper No. 16080 Page 5 of 9
Table 1. The most relevant features
#
Feature
Description
Classification
accuracy
1
O2---fft_rel_gamma_(30_to_40Hz)
Relative power in gamma waveband
for channel O2
0.7602
2
coher-theta---O1-O2
Coherence in theta waveband between
channels O1 and O2.
0.8160
3
coher-theta---T4-O2
Coherence in theta waveband between
channels T4 and O2
0.8473
4
T3---fft_rel_delta_(1_to_4Hz)
Relative power in delta waveband for
channel T3
0.8575
5
O1---fft_rel_(15_to_16Hz)
Relative power in gamma waveband
for channel 01 (15-16Hz)
0.8633
6
coher-beta---O1-O2
Coherence in theta waveband between
channels O1 and O2
0.8660
7
coher-gamma---T3-O1
Coherence in theta waveband between
channels T3 and O1
0.8745
8
O1---fft_rel_(3_to_4Hz)
Relative power in 3-4 Hz for channel
O1
0.8825
9
T3---fft_abs_(39_to_40Hz)
Absolute power in 39-40 Hz for
channel T3
0.8879
10
coher-theta---Fp1-Fp2
Coherence in theta waveband between
channels Fp1 and Fp2
0.8919
Based on this selection, it was possible to reduce the number of electrodes used. Later measurements only used 6
EEG channels: Fp1, Fp2, T3, T4, O1, and O2. This simplified the process of the application of the EEG device and
also later data processing.
After the first phase of EEG data analysis, we were able to create a so-called mental workload index that is based on
the classification model. The first index, I1, has a value in the interval of <0-1>, where 0 is at maximum rest and 1 is
at maximum mental workload. Index I2 can more effectively differentiate the phases of mental workload because its
value is (0, 1). The index is based on the statistical analysis K-NN (k-nearest neighbors) that finds the maximum rest
phases (0) and the phases with maximum mental workload (1). An example of index I1 is presented below (see
Figure 3).
Interservice/Industry Training, Simulation, and Education Conference (I/ITSEC) 2016
2016 Paper No. 16080 Page 6 of 9
Figure 3. Index I1 showing mental workload (blue lines) during demanding situations (black boxes)
To ensure we obtained the most relevant data, we used a triangulation of the methods, so we also added ECG
measurement to the study. An analysis of ECG data regarding heart rates was made to show the ability of this
method to identify mental workload. The resting phase was compared to the average heart rate during flight, and a t-
test showed significant differences between the two phases at a significance level of p < 0.01. To easily work with
the data obtained, we created a heart rate coefficient (HRC) that is based on the ratio to the resting HR phase. The
bigger this HRC is the higher was the mental workload experienced by the pilot tested (see Figure 4). This also
includes the pilot’s own subjective reference from the post-flight questionnaire (grey boxes).
Figure 4. Heart rate coefficient (blue) increases in demanding situations (boxes)
Thanks to the relative simplicity of the ECG measurement, the instructor can decide right after the measurement is
taken if there were some parts of the flight that were mentally demanding and require further training. The pilot’s
physiology is divided into zones regarding the HRC. If the HRC didn’t exceed 1.3 during the flight, then the pilot
didn’t pay full attention, which was reflected in the mistakes made that in some cases led to the pilot being shot
down. If the HRC was between 1.3-1.5, then the pilot was in the optimal state, fully focused and delivering the best
performance. If the HRC was higher than 1.5, then the pilot was experiencing an overwhelming mental workload
which resulted in mistakes, bad decision making, rash reactions and inaccurate communication. These conclusions
are based on an instructor evaluation of the pilot’s performance. The summary of these results are shown in Table 2.
Interservice/Industry Training, Simulation, and Education Conference (I/ITSEC) 2016
2016 Paper No. 16080 Page 7 of 9
Table 2. Level of mental workload
HRC
Level of mental workload
1-1.3
Lack of attention
1.3-1.5
Optimal state
> 1.5
Overload
Later, the questionnaires were assessed to compare the mental workload obtained from the EEG and ECG devices to
the subjective perceptions of the pilots. As mentioned above, the questionnaire had two parts. In the first part
regarding social support and the current mental and physical states, none of the pilots scored in extremes, and all of
them showed results from numbers 1-3. So, this should not be a variable that could interfere with the study. The
second part was focused on mental workload and perceived difficulty during specific demanding situations. Given
the instructions, the pilot was asked to imagine the most mentally demanding situation that anyone can experience
and scored 20 to this event. Later, he was asked to imagine his personally most demanding event and again scored
that on a scale of 1-20. From then on, all of the given situations that were to be judged were also on this 1-20 scale,
so the subject could compare his biggest personal level of mental workload to that usually experienced during the
flight task. Results from this questionnaire show that the most demanding situations were Threatcall, Crossing Flot,
and unexpected situations like system failures (radar interference, losing connection with GCI).
The evaluation questionnaire showed the results of the subjects’ performance in difficult situations and also the
overall rating from the instructor for the whole mission. To verify the use of the subjective reference questionnaire
about mental workload, it was correlated with the evaluation questionnaire. The aim was to find out if the use of the
self-reporting questionnaire is sufficient for detecting the most mentally demanding situations. The results didn’t
show a significant correlation (p < 0.01) between the mental workload reported by the pilot and the evaluation by
the instructor. This output emphasized the need for using more relevant methods not linked to subjective reference.
Subjects in this study tended to underestimate the level of mental workload perceived and its impact that appeared in
the later evaluation made by the instructor.
Discussion
The aim of the study was to test the sensitivity and usefulness of psychophysiological measures of mental workload
during diverse segments of simulated tactical combat missions. The psychophysiological measures like EEG and
ECG proved to be very efficient methods in detecting the mental workload of military pilots during simulated flight.
More specifically, our experiment showed that the relative power in the gamma waveband for channel O2 was
highest during the most demanding situations. This finding is quite innovative, because the gamma waves haven’t
yet been considered in any other flight research. Also, changes in the heart rate, as measured by ECG, appeared to
reveal the most demanding situations during flight. The course of the mean heart rate may be strongly related to
emotional factors, such as danger during flight segments. However, these may be influenced by information
processing. Our study reliably showed that heart rate values were greater than those found for resting on the ground.
Conclusion
The conclusions drawn, from this proven usefulness of psychophysiological measurements, were used to measure
the in-flight mental workload of military helicopter pilots in the first stages of training. The implementation of the
complex system is still in progress, but the ongoing results are demonstrating the potential for helping instructors to
detect the more difficult parts of the flight that should be more extensively trained, even when it has been technically
well-handled. If the maneuver was made correctly but it was too mentally demanding, it should still be trained on
further to gain more confidence for the pilot and to thoroughly complete the learning process. To make the
measurement more user-friendly, we developed an application that evaluates the psychophysiological data for the
instructor who can later decide about further training. It’s a hands-on tool where the instructor downloads data from
the ECG measurement and can later review the most demanding parts of the flight. His final decision is based on the
heart rate coefficient that is divided into zones of mental workload that occurs during flight. It there were some parts
of the flight where the pilot wasn’t in the optimal state, the instructor focused on this specific task in follow-up
training. Due to recurring practice of the task, pilot tension decreased. This was reflected in the physiological data
and in better performance. Specific data aren’t yet available due to ongoing research and its resulting application.
The data from the EEG measurement require more elaborate analysis because of the artifacts and other variables
present. So, it is necessary to have specialist present for measurement and evaluation.
Interservice/Industry Training, Simulation, and Education Conference (I/ITSEC) 2016
2016 Paper No. 16080 Page 8 of 9
Our research has improved the awareness of the fundamental role of training within aviation that can be enhanced
by the detection of mental workload. Due to the significance of the methods used, we were able to detect the most
mentally demanding situations during real or simulated flight activity. Further analysis showed some individual
differences that led us to the future development of individual norms for each pilot. Psychophysiological methods
are nonintrusive, and modern recording equipment does not interfere with pilot performance. So, it appears to be
about time they are included in every aviation training program focused on increasing mental resilience.
REFERENCES
Bayevsky, R. M., Ivanova, G. G., Chireykin, L. V., Gavrilushkin, A. P., Dovgalevsky, K. U., Mironova, T. F., et al.
(2002). HRV Analysis under the usage of different electrocardiography systems (methodical
recommendations).
Borghini, G., Astolfi, L., Vecchiato, G., Mattia, D., & Babiloni, F. (2014). Measuring neurophysiological signals in
aircraft pilots and car drivers for the assessment of mental workload, fatigue and drowsiness. Neuroscience
& Biobehavioral Reviews, 44, pp. 58-75.
Boucsein, W. (2012). Electrodermal Activity. New York: Springer Science & Business Media.
Budzynski, T., Budzynski, H., Evans, J., & Abarbanel, A. (2009). Introduction to Quantitative EEG and
Neurofeedback: Advanced Theory and Applications. Elsevier.
Dahlstrom, M., Nahlinder, S., Wilson, G. F., & Svensson, E. (2011). Recording of Psychophysiological Data. The
International Journal of Aviation Psychology, pp. 105-122.
Di Stasi, L. L., Antoli, A., & Canas, J. J. (2011). Main sequence: An index for detecting mental workload variation
in complex tasks. Applied Ergonomics (42), pp. 807-813.
Dussault, C., Guezennec, C., & Jouanin, J. (2004). EEG and ECG Changes During Selected Flight Sequences.
Aviation, Space, and Environmental Medicine, 75(10), pp. 889-897.
Friedl, K. E., Grate, S. J., Proctor, S. P., Ness, J. W., Lukey, B. J., & Kane, R. L. (2007). Army research needs for
automated neuropsychological tests: Monitoring soldier health and performance status. Archives of Clinical
Neuropsychology(22), pp. 7-14.
Hart, S. G., & Hausser, J. R. (1987). Inflight application of three pilot workload assessment techniques. Aviation,
Space, and Environmental Medicine, pp. (58), 402-410.
Jacobs, G. D., & Friedman, R. (2004). EEG spectral analysis of relaxation techniques. Applied Psychophysiology
and Biofeedback, pp. 245-254.
Jorna, P. (1993). Heart rate and workload variations in actual and simulated flight. Ergonomics, 36, pp. 1043-54.
Kakimoto, Y., Nakamura, A., & Tarui, H. E. (1988). Crew workload in JASDF C-1 transport flights: Change in
heart rate and salivary cortisol. Aviation, Space, and Enviromental Medicine, 59, pp. 511-516.
Karthikeyan, P., Murugappan, M., & Yaacob, S. (2013). Detection of human stress using short-term ECG and HRV
signals. Journal Of Mechanics in Medicine & Biology, 13(2), p. 1.
Kubota, Y., Sato, W., Toichi, M., Murai, T., Okada, T., & Hayashi, A. (2001). Frontal midline theta rhythm is
correlated with cardiac autonomic activities during the performance of an attention demanding meditation
procedure. Brain Research. Cognitive Brain Research, pp. 281-287.
Lal, S., & Craig, A. (2002). Driver fatigue: Electroencephalography and psychological assessment.
Psychophysiology, pp. (39) 313-321.
Rauch, H. H., Karpul, D., Derman, W., & Prinsloo, G. (2013, 38 (1)). The Effect of a Single Session of Short
Duration Heart Rate Variability Biofeedback on EEG: A Pilot Study. Applied Psychophysiology &
Biofeedback, pp. 45-56.
Sokhadze, E. T. (2012). Peak Performance Training Using Prefrontal EEG Biofeedback. Biofeedback, pp. 40 (1), 7-
15.
Thomas, M. L., & Russo, M. B. (2007). Neurocognitive monitors: Toward the prevention of cognitive performance
decrements and catastrophic failures in the operational environment. Aviation, Space, and Environmental
Medicine (78), pp. 144-152.
Tracey, I., & Flower, R. (2014). The warrior in the machine: Neuroscience goes to war. Nature Reviews
Neuroscience (15), pp. 825–834.
Veltman, J. A., & Gaillard, A. W. (1993). Indices of mental workload in a complex task environment.
Neuropsychobiology, pp. 28(1-2), 72-75.
Interservice/Industry Training, Simulation, and Education Conference (I/ITSEC) 2016
2016 Paper No. 16080 Page 9 of 9
Wilson, G. F. (2002). An analysis of mental workload in pilots during flight using multiple psychophysiological
measures. The International Journal of Aviation Psychology, 12, pp. 3-18.
Yi, B., Matzel, S., Feuerecker, M., Hörl, M., Ladinig, C., Abeln, V., et al. (2015). The impact of chronic stress
burden of 520-d isolation and confinement on the physiological response to subsequent acute stress
challenge. Behavioural Brain Research (281), pp. 111-115.
