Philippe Fauquet-Alekhine, Th.

Geeraerts
and Laetitia Rouillac
Characterization of anesthetists’ behavior
during simulation training: performance
versus stress achieving medical tasks with
or without physical effort
Article (Published version)
Refereed

Original citation:
Fauquet-Alekhine, Philippe, Geeraerts, Th. and Rouillac, Laetitia (2014) Characterization of
anesthetists’ behavior during simulation training: performance versus stress achieving medical
tasks with or without physical effort. Psychology and Social Behavior Research, 2 (2). pp. 20-28.

DOI: 10.12966/psbr.06.01.2014

Reuse of this item is permitted through licensing under the Creative Commons:

© 2014 The Authors

CC BY 3.0

This version available at: http://eprints.lse.ac.uk/68664/

Available in LSE Research Online: December 2016

LSE has developed LSE Research Online so that users may access research output of the
School. Copyright © and Moral Rights for the papers on this site are retained by the individual
authors and/or other copyright owners. You may freely distribute the URL
(http://eprints.lse.ac.uk) of the LSE Research Online website.
Sciknow Publications Ltd. PSBR 2014 2(2):20-28
Psychology and Social Behavior Research DOI: 10.12966/psbr.06.01.2014
©Attribution 3.0 Unported (CC BY 3.0)

Characterization of Anesthetists’ Behavior during Simulation

Training: Performance Versus Stress Achieving Medical Tasks
with or without Physical Effort
Fauquet-Alekhine1,2,3,*, Geeraerts4, Rouillac2
1
Nuclear Power Plant of Chinon, BP80, 37420 Avoine, France
2
Lab. for Research in Science of Energy, Montagret, 86200 Neuil ss Faye, France
3
Inst. of Social Psychology, London School of Economics and Political Science, Houghton St., WC2A 2AE, London, UK.
4
Centre Hospitalier Universitaire, Pole Anesthésie-Réanimation, Place du Dr Baylac, 31059 Toulouse, France
*Corresponding author (Email: p.fauquet-alekhine@lsc.ac.uk)

Abstract - Decades of research about stress have shown that it could be source of performance but also of cognitive deficit. The
studies have led to highlight occupational stress variables that researchers have characterized by physiological measurements,
data treatments and protocols becoming more and more complex with time. If these devices are gaining in precision, they are
now too complex to allow non-specialist users to produce a quick interpretation of results. Yet for vocational training,
specifically on simulators, trainers need to know in real time whether or not what they implement allows the trainees to learn in
good conditions, i.e. by favoring the behavior produced by the positive effect of stress on performance. The present paper
addresses the performance versus occupational stress during training sessions of anesthetists on simulator. We studied the
performance and stress with or without physical effort using a simple protocol based on the use of basic heart parameters in order
to obtain a quasi-instantaneous interpretation of the data. We identified cognitive deficit zone during training according to the
Yerkes & Dodson (1908) relationship between performance and stress. We showed that performance versus stress during
simulation training with or without physical efforts could be successfully analyzed for immediate assessment of stress
influencing performance. Suggestions have been made for improving training sessions and avoid trainees’ behavior induced by
cognitive deficit. Limits of the protocol are exposed.
Keywords - Performance, Stress, Simulation

performance have already been obtained by others (for

1. Introduction example see Broadhurst, 1957; Drach-Zahary & Erez, 2002;
Hancock et al., 2002; and the review of Staal, 2004, Pearsall
For many years, stress has been shown by researchers to be et al., 2009).
both a source of performance and a source of cognitive But for some industrial contexts concerned by the impact
deficits, therefore reducing performance. Getting information of stress on performance, such a complex organization cannot
about the kind of influence of stress in a work activity has be applied, for a matter of time, money, and competencies,
appeared very useful to be able to work mainly under the while it would be of great interest to have better knowledge in
positive influence of stress. For this aim, qualitative particular cases: classic training session, training on simulator,
considerations help, but the best is to maintain quantitative evaluation, crisis management… To address this need, we
approach because of the objectivity. Many works have been tested the use of a simple protocol, requiring basic metrology
done in order to make the link between stress and and simple straight data analysis on training simulators with
physiological parameters in a quantitative approach and trainers who are not necessarily experts in medical researches.
studies have shown how to measure some of those parameters This does not mean that we aim at demonstrating that
identified to be closely associated with the stressed state of protocols involving the stress description through a full set of
subjects. Nowadays, with technical progresses and after physiological parameters (such as breathing rate, heart rate
conclusions of several decades of studies, medical facilities variability, hormone excretion, inter-beat interval consistency,
are available to do such investigations, which require specific blood pressure modulation, etc.) must be put away. We just
devices, metrologies, and then demand specific software for suggest something else: a performance vs stress approach
analysis. At each step, specialists are necessary. Different using a simple protocol and device for straightforward
relationships between the occupational stress and the
 Psychology and Social Behavior Research (2014) 20-28 21

analysis. variability analysis, combined with respiration analysis and

sudation. As said above, these techniques involve
1.1. Stress types sophisticated metrologies and elaborated software which need,
The stress which we were interested in this study is a short thereafter, a careful data examination to be sure of the
term stress, compared to long term stress linked with chronic conclusions (Fairclough et al., 2004; Montano et al., 2009;
stress exposure (refer for example to the studies of Maslova et Müller et al., 2009; Rohleder et al., 2009; Schubert et al. 2009;
al. (2002) who studied the effect of chronic stress on arterial Bailon et al., 2010). Using these techniques, Schubert et al.
blood pressure, or studies of Schubert et al. (2009) who (2009) (as others before: Steptoe, 2000) have confirmed again
compare both kinds of stress). All types of stress are how the mean heart rate increases with short term stress while
concerned by psychological and physiological variable it decreases with long term stress, showing that, for the kind of
changes (McLean, 1974; Beehr & Newman, 1978; Karasek & stress we were interested on, the mean heart rate is a pertinent
Theorell, 1990; Palmer et al., 2003), some of them linked with parameter to characterize subjects’ state of stress.
genetics (Judge et al., 2012). In case of long term stress, Aiming at elaborating a simple and straightforward
physiological parameters vary differently than in case of short protocol for performance vs stress analysis, we focused on
term stress. The time component is fundamental and specific. heart rate measurements as physiological parameters to
For the short term stress, time is mainly relevant in terms of qualify stress.
length, while for long term stress, time is relevant in terms of
frequency, which may lead to ―chronic stress‖. 1.3. Short term mental stress with or without physical effort
As we considered here stress at work, the stress was of According to the bibliographic research, no publication is
occupational kind. We thus studied short term occupational available concerning the analysis of stressful work situations
stress, at which people at work are submitted when they are involving subjects with a comparative study of the conditions
asked to perform a task bounded in a short time interval (about thereafter said with or without physical effort.
several seconds to several hours) rather than stress itself. We Our contribution specifically focuses on this comparative
studied here the relationship between performance and stress, point.
and mainly the influence of the conditions of stress on the
performance. Yerkes & Dodson (1908) gave a theoretical
description of this relationship, assuming that performance 2. Materials and Methods
rises with the stress level until a given threshold beyond The method was based on the evaluation of the subjects’ state
which performance decreases, suggesting that stress puts the
of stress using perception questionnaire and heart rate
subject in a cognitive deficit zone.
measurements combined to the assessment of the
performance. These quantitative data were supplemented by
1.2. Heart parameters measurements
qualitative material obtained through interviews during
Measurement of the physiological parameters helps the debriefings of training sessions. The experiment was applied
characterization of the subject’s state with regards to the stress. to anesthetist training sessions on full scale simulator.
Here, a short recall of the physiological process is done. We
shall not remind here the details of the physiological process 2.1. The characterization of stress conditions by perception
taking place with occurrence of stress. We shall just focus on questionnaire
the main lines in order to justify the choice of the Several questionnaires for self-rating of stress have been
physiological parameters measured in our experiments. established and scientifically tested. The Job Content
To make it short, we shall consider a stressed subject in a Questionnaire of Karasek has not been retained here because,
given context. In case of an induced psychological pressure even if the macro-variables are watched through the items, a
for the subject, two main systems will be called upon: the lot of questions not concerned by the training sessions are
sympathetic nervous system and medulo-surenal gland, asked and some variables which are relevant to be asked are
producing adrenaline / noradrenaline and corticotropin not investigated by the questionnaire. The Cohen’s Perceived
releasing hormone (CRH). The first one will have short term Stress Scale (PSS) (Cohen et al., 1983) as the more recent
effects, mainly concerning a rise in blood glucose, heart and Work and Well-Being Questionnaire (Kilminster et al., 2007;
ventilation rate. The second one will have a long term effect Bridger et al., 2011), concern the long term stress and thus is
(at least several hours), mainly related to the production of not adapted to this study. The State-Trait Anxiety Inventory
corticotropes and cortisol which have a moderating effect as (especially the STAI form Y-A self-rating the subject’s
opposed to adrenaline. The short term stress is thus mainly anxiety state) developed by Spielberger (1983) has not been
concerned by the sympathetic nervous system and its effects used because it measures anxiety with too few reference to
(see for example Davezies, 2008; Montano et al., 2009; Keitel exogene parameters. The perception questionnaire of stress
et al., 2011). used for this experiment is the Peritraumatic Distress
Lots of studies are available concerning physiological Inventory (PDI; 13 items). It has been elaborated in order to
measurements associated with state of stress of the subjects. obtain a quantitative measure of the level of distress
Many publications are available about the spectral heart rate experienced during and immediately after a traumatic event.
22 Psychology and Social Behavior Research (2014) 20-28

This questionnaire has been scientifically tested by several specialization, observed on full scale simulator.
(see for example Brunet et al., 2001), including in its French Students were involved in a one day training session in
form (see Jehel et al., 2005 and 2006). It presents the operating theatre, and training was performed the whole week
advantage, compared to the STAI, to include items such as the (5 days). It means one different group of about 6 students was
frustration or guilt not to do more, the shame, the fear for received every day. At the end of the week, 27 French
one’s safety or for others, which are important parameters students were trained, playing different role depending on the
concerning the job. It includes also the subject’s feelings scenario.
concerning physiological parameters (sweating, shaking, Four different scenarii were used per day (less than one
pounding heart). The problem for this questionnaire is that it hour each), and 3 students were training together per scenario,
is linked with the diagnosis of posttraumatic stress disorder each scenario (about 30 min) followed by a debriefing session
(PTSD) which requires that a subject has high levels of (30 to 45 minutes).
distress during or after the traumatic event. We shall see The participants of the simulated situation for a scenario
thereafter that it can be a drawback when the subject is were:
submitted to a too low level of stress. - 3 students playing the role of physician, nurse, and
help,
2.2. Physiological measurements during training - 1 physician trainer, playing the surgeon,
On the contrary of sophisticated metrologies and elaborated - 1 physician trainer piloting the simulator.
software which need, thereafter, a careful data examination to This implied that the number of cases available depended
be sure of the conclusions (Montano et al., 2009; Rohleder et on the students who were involved in the situations and on the
al., 2009; Schubert et al. 2009; Bailon et al., 2010; De role they played. At the end of the week, the sample of
Jonckheere et al., 2010; Jo et al., 2013), we aimed at a simple subjects concerning the actor physicians was (N=18; 50%
solution based on heart rate. male), and the sample of subjects concerning the actor nurses
Heart rate (HR) has been measured using a Polar FS2c was (N=18; 44.4% male).
composed of two parts. The first one is a detector with two Deontology has been presented during each introduction
electrodes to be put on the breath, touching the skin, close to of the Stress-test or training sessions with the subjects. First of
the heart. The second one is a monitor which looks like a all, all subjects were volunteers. It was explained that all data
watch which can be worn on the wrist. The screen shows the would be used for research, anonymously, and that no access
mean value of the measured heart rate. The whole device is to personal data or to the links between data and identity
worn by the subject and at the end of the test, values of the would be given to anyone. A specific form was filled up and
mean heart rate and of the maximum are given. The technical co-signed by the subject and the researcher each time.
specifications are: Concerning physiological parameters, it was only
- accuracy of time measurement: better than  2.0 s / measured for the student playing the physician’s role: heart
24 h rate was recorded by the Polar monitor described above, and
- accuracy of heart rate measurement:  1% or  1bpm, gave mean and max values at the end of the session.
whichever larger During the simulated situation, the other students watched
- measuring range : 15-240 bpm a video projection of the simulation in another room, together
It has been found (Fauquet-Alekhine et al, 2011 & 2012) with the researcher and other physician trainers. In this room,
that, to have pertinent heart parameters concerning stress, we a large size screen and an audio device allowed to watch and
could use a mean value and a maximum value of HR. We thus hear what was going on in the operating theatre.
chose to use the basic parameters which are the mean heart The general pedagogical goal was to put students in a
rate (HRmean) and maximum heart frequency (HRmax) to simulated situation pertinent to their future job where they
characterize the subjects’ state of stress, keeping in mind that need to make diagnosis, take decision and act to deal with the
both parameters increase with stress intensity in the case of critical case.
the short term occupational stress. The scenarii were clinical cases involving only one
In addition, samples of salivary amylase were taken from dysfunction (no cumulative cases).
each subject involved in the anesthetist’s training just before The 4 scenarii were:
and just after being involved in the simulated situation. The - Asphyxia related to post-operative cervical
aim was to analyze the evolution rate (not discussed in this hematoma,
paper). - Local Anesthetics intoxication,
- Peroperative third degree auriculoventricular block,
2.3. Subjects sample, training context and pedagogical - Peroperative respiratory arrest related to injection of
goals myorelaxant drug.
All the presented experiments have been conducted with and For further details about scenario contexts, see Geeraerts
within a hospital university of Paris district. The subjects for et al. (2013).
this application were French students, all residents in
Anesthesiology and Intensive Care in their third year of
 Psychology and Social Behavior Research (2014) 20-28 23

2.4. Performance evaluation and link with stress for a short mental occupational stress (Fauquet-Alekhine et al.,
Results presented in a previous work (Fauquet-Alekhine 2011). The determination coefficient of polynomial fitted
et al., 2011 and 2012) demonstrated that a Yerkes & Dodson curve is R2 = 0.69.
curves could be fitted for performance concerned with short Using the concept of Human Functional States (HFS)
mental occupational stress. In this previous work, suggested by Leonova (2009), we divided the bell curve into
experiments were carried out with healthy subjects (N=18; 50% three main Human Functional States (Fauquet-Alekhine,
male), about 25-35 yo., same academic background and the 2012):
same kind of job, taking a Stress-test made up of 12 simple - the left part is linked to a HFS of positive state of
questions. A 100% success could be expected for all of the stress or stable cognitive state, where performance
subjects taking the test but we can see that it was not the case: rises with the stress,
a modal analysis of the performance coefficient Kp (Fig. 1) - the central part reflects a HFS of transience (transient
has shown that there was a range over which values spread. state) for the subject in terms of stress effects, where
The range was yet narrow. performance has raised with stress until a given
threshold beyond which the variation is inverted,
- the right part concerns a HFS of negative state of
stress or potential cognitive deficit state, where stress
tends to put the subject in a cognitive deficit state,
reducing the subject’s capacity to fully use his/her
cognitive resource and making performance
decreasing.
These HFS are drawn on the graph presented in Fig. 3.

Fig. 1. The modal distribution of Kp for Stress-test.

Analysis of measurements of heart rate HR
(discrimination of conditions, modal analysis) and correlation
with results of the perception questionnaire of stress have
allowed us to define a reduce coefficient of stress Ksr as :
Ksr = HRmean . HRmax ampl
where HRmax ampl = HRmax - HRmean
Plotting the subjects’ performance measured through the
performance coefficient Kp vs the state of stress rated by the
reduced stress coefficient Ksr, we obtained (Fauquet-
Alekhine et al., 2011 and 2012) the bell curve (Fig. 2) where
subjects working in stressful conditions were well
discriminated on the right side of the graph (clear squares)
from other subjects. Fig. 3. Human Functional States (HFS) divided into three
1 main parts: i) central part: transient state for the subject in
terms of stress effects, ii) left part: positive state of stress, iii)
0,8 right part: potential cognitive deficit state.
R²= 0,693
0,6 The coefficients of performance Kp vs stress Ksr gave a
good representation of the subject’s state of performance vs
Kp

0,4 stress during a work activity in which mental stress is

involved, which means without physical stress. This method
0,2 has been applied in the present study, and extended to the case
of short term occupation mental stress with significant
0
physical effort.
0 1000 2000 3000 In the present work, all training sessions were video
Ksr
recorded in the aim of performance rating. All records have
Fig. 2. Experimental data obtained during the Stress-test, been carefully watched by anesthetist-trainers afterwards in
plotted with performance coefficient Kp vs reduced stress order to identify two specific times concerning the reach of
coefficient Ksr and fitting a Yerkes & Dodson (1908) curves expected results for each scenarii, that is the length of time
taken by the anesthetist in the simulated situation to identify
24 Psychology and Social Behavior Research (2014) 20-28

the symptoms, perform the etiologic diagnosis, take the 3.3. Performance measurement
symptomatic corrective measures, and take the etiologic To evaluate performance of the students, we used time
corrective measures. The selected times for performance measurements concerning the reach of the final result:
assessment were thus defined: - Time to perform the symptomatic corrective
- Time to perform the symptomatic corrective measures: tsc.
measures: tsc. - Time to perform the etiologic corrective measures:
- Time to perform the etiologic corrective measures: tec.
tec. From these two parameters evaluated on the basis of
The length of time taken to identify of the symptoms or analysis of video records of the sessions, we could build a
perform the etiologic diagnosis were not selected because we performance coefficient Kp. It was based on the following
observed that, in case of low performance, the delay between remarks induced by the basic postulate that performance is
decision and action could be very long. linked with the right result (the problem is correctly solved) in
the minimum of time.
From this postulate, we suggested that a relevant Kp
3. Results decreases if time of observation or action increases: the
3.1. Quantitative data from measurements students are less efficient if they take more time to deal with
the problem.
Two kinds of data were available for actor physicians only:
The formulation is thus:
- Mean and maximum heart rates.
- Kp increase with inverse of time corrections tsc and
- Evolution rate of salivary amylase (not discussed in
tec.
this paper).
From this postulate, we built the coefficient as the sum of
A previous analysis (Geeraerst et al, 2013) showed that
the inverse of tsc and tec:
data had to be considered separately: on one hand, trainees
Kp = 1/tsc + 1/tec
involved in a significant physical effort and on the other hand
The Kp has been tested on simulated data with good
for other trainees. If we considered in the whole set of data all
agreement to expectation. The analysis of Kp with a limit
cases of subjects, the resulting cluster could not be explained
approach confirmed the reliability of the Kp. The modal
in terms of performance versus stress: the data did not show
distribution of Kp for anesthetists’ training sessions is
any specific shape, the bell curve did not appear and no other
presented on Fig. 4 according to the following modes:
specific shape as well.
[-inf; 0,001[, [0,001; 0,0075[, [0,0075; 0,01[, [0,01; 0,015[,
Two subjects did cardiac massage which induces a [0,015; 0,02[, [0,02; 0,025[, [0,025; 0,03 [, [0,03; +inf[
significant physical effort and a disturbance of HR. Others We can see that the range over which values spread is
implicated in the first scenario were confronted with a larger than for the Stress-test (Fig. 2). This is due to the fact
difficult intubation case. This also involved significant that for the Stress-test, subjects were expected to reach a 100%
physical efforts as we have observed in situation. The data score as the questions of the test were easy, while for these
related to these cases were treated separately from the others. training sessions, subjects discover the situation within an
The comparison of the HR values have shown that HR mean initial training and so are not expected to get a result tending
covered the same range of values for both cases (with and to 100% success.
without significant physical effort), and that HR max ampl
covered a narrower range of values in the case of significant
physical effort.

3.2. Perception through PDI questionnaire of stress

The questionnaire has been filled in by the subjects
immediately after taking the test. To use the results, we
affected to the answers a value from 0 to 4 according to the
Likert scale suggesting: not at all, a few true, rather true, very
true, extremely true. Then we calculated from the 13 answers
for each subject a mean value, Qmean.
If the Cronbach’s alpha calculated was rather good ( =
0.8), the data related to the reduced stress coefficient Ksr Fig. 4. The modal distribution of Kp for anesthetists
versus Qmean gave a correlation coefficient rather bad training sessions.
(Geeraerst et al., 2013). Yet the set of dots showed a cluster
3.4. The case of anesthetists’ training without physical
with a coherent global increase of both parameters. The low
efforts
value of the correlation coefficient may be due to a poor
differentiation of low stress cases and to an over Considering only the cases of simulated situations with no
self-estimation of stress in stressful cases. disturbance due to physical effort, the performance coefficient
Kp built as described in previous section vs the reduced stress
 Psychology and Social Behavior Research (2014) 20-28 25

coefficient Ksr gave a distribution of points of Yerkes & deficit zone on a Kp vs Ksr graph of Yerkes & Dodson type
Dodson curve type (Fig.5) and led to the following curve (see Fig. 5 and 6), on the right side of the bell graph,
conclusions: most of the students were actually in the both for work activities with or without significant physical
cognitive deficit state according to the Human Functional efforts.
States (HFS) defined above (Fig. 3). After analyzing the context of training on the basis of
training sessions debriefing with trainees, the conclusions
were that residents needed to be more familiar with the
simulator and with the activity before being involved in this
kind of working situations. Thus the main point of
improvement would be to make them familiarized with the
simulator before the training session itself, with a progressive
approach of the simulator in several steps distributed on
several days, including the familiarization with observers
whilst working on simulator.
Another point of improvement according to the debriefing
interviews was that trainees had to be able to perceive their
knowledge and skills sufficient for the task in the perspective
of increasing self-confidence: this implies to create or manage
differently the previous steps of their training.
Fig. 5. Performance coefficient Kp versus reduced stress
coefficient Ksr for anesthetist residents not submitted to strong 4.2. Influence of experienced simulation
physical efforts. Furthermore, another interesting point appeared. Checking
the cases of students who had already been involved in at least
3.5. The case of anesthetists’ training with physical efforts
one simulation training before taking the present experimental
Considering only the students concerned by occupational session, data showed for both cases (concerned or not by a
stress including a significant physical effort, we plotted the Kp significant physical effort) less stress for these students than
versus Ksr. We obtained the following graph Fig. 6. for the students who discovered the simulator. This confirmed
previous analysis (Geeraerst et al., 2013) which pointed out
that this kind of experience influences stress but not
performance of the trainees. Thus, it demonstrated that a
progressive training would be benefic and that being used
with this stress variable may reduce its influence as a stressful
contributor.

4.3. The comparative state of stress with or without

physical effort
As shown on the above graphs, the performance vs stress data
could be fitted with bell curves, both in the case of residents
who were involved in a significant physical effort during the
Fig. 6. Performance coefficient Kp versus reduced stress work activity and for these who were not. Superimposing on
coefficient Ksr for anesthetist residents submitted to strong the same graph these curves named NPE for ―No Physical
physical efforts. Effort‖ and WPE for ―With Physical Effort‖ (Fig. 7), we
pointed out several findings:
In this case, we recognize again a Yerkes & Dodson curve - The WPE curve was higher than the NPE curve in
type with most of the data located in the HFS of negative state terms of Kp, which means that the residents’
of stress or potential cognitive deficit state. performance was higher when involved in a
The conclusions were that most of the students were significant physical effort while dealing with the
actually in the HFS of negative state. operation. Checking the experienced status of the
subjects (already trained on simulator or not) showed
no bias as experienced subjects were involved in
4. Discussion both cases.
4.1. The stressful aspect of the training and the factors of - The WPE curve was larger than the NPE curve in
improvement terms of Ksr by the left side of the X-axis, which
Application of the developed protocol studying performance means that stress of the residents involved in a
vs stress has confirmed stressful conditions for anesthetists’ significant physical effort covers lower values. We
training, showing that most of the subjects were in a cognitive checked whether any physiological characteristics of
26 Psychology and Social Behavior Research (2014) 20-28

the subjects could produce these special shapes of teams on simulators, then it might contribute to explain the
curves but we could not find any bias. higher values of Kp for the WPE cases than for the NPE cases.
- The WPE and NPE curves went along the same From the physiological standpoint, the relaxing effect of
decreasing values in the HFS of potential cognitive endorphins (released into blood while physical efforts)
deficit which means that in this zone, the relationship modifying HR values has been rejected because this effect
between performance and stress did not depend on occurs several tens of minutes after the beginning of the effort.
the subject’s involvement in a significant physical In our cases, the simulated situations were too short.
effort within this HFS. Another explanation may be suggested regarding
- The extreme value of Kp of the curves were related to performance. As the possible bias due to subjects’ training
different values of Ksr; this corresponds to two experience was eliminated, the higher performance for
different thresholds of optimized stress state. It subjects submitted to a significant physical effort (physical
appeared that for the WPE curve, this threshold stress) might be explained according to the following
occured sooner in terms of stress than for the NPE hypothesis: facing the emergency of an action that will save
curve. It could be seen as a drawback, but at the same the patient (cardiac massage or strong intubation), the subject
time the HFS were larger, offering thus a larger range was forced to make a decision faster that in the other cases.
of values of Ksr for which performance grew up. The subject had less time to think about different questions, to
doubt. The decision was therefore made earlier and increased
0,035
thus the performance coefficient.
0,03
NPE
0,025
0,02
WPE 5. Conclusions
Kp

0,015 Demonstration was made for performance assessment versus

0,01 short term mental occupational stress using a simple protocol
0,005 and device. The results obtained in experimental and
0 industrial conditions matched the Yerkes and Dodson theory
0 2000 4000 6000 8000 (1908).
Ksr The simple protocol and device consist of:
- for each subject, the determination of a stress
Fig. 7. Fitting curves for the two data cases named NPE coefficient or a reduced stress coefficient on the basis
for ―No Physical Effort‖ and WPE for ―With Physical of the mean and maximum heart rates measured by
Effort‖. an individual portable heart rate meter during the
Explanations of these findings may be as following. work activity,
When subjects were submitted to a significant physical - for each subject, the determination of a standard
effort (physical stress), the difference between HR mean and performance coefficient according to the work
HR max was less important than when the subjects were activity.
submitted only to a mental stress. The values of HR mean and The simple protocol and device qualified previously
HR max ampl were compared with regards to the cases WPE and (Fauquet-Alekhine et al., 2011 & 2012) and used here can be
NPE. In particular, for HR max ampl the range for WPE was successfully applied to:
narrower than for NPE: in bpm, about [10; 30] for WPE and - a group of subjects involved in the same work
about [15; 50] for NPE. This was probably due to the physical activity,
effort making the mean HR more close to its extreme value. - a case of performance assessment vs short term
Thus, HR max ampl was less important with physical stress and mental stress which means that the time length of the
consequently the coefficient of stress Ksr too. This produced activity must be short (several tens of minutes max)
an enlargement of the HFS and a shift to lower values of the with or without significant physical efforts for the
extreme of the curve. This may be combined to another effect: subjects if it is possible to consider these two cases
when the subject was involved in an action requiring separately.
significant physical effort, the subject focused on this action The results and discussion presented in this paper showed
which lessened the influence of other stressful factors and that:
thus contributed to reduce their effect. We therefore might - The subjects (French healthy anesthetist residents
suggest the assumption that the subject was less stressed. trained on full scale simulator) could be divided in
Concerning the higher values of Kp for the WPE cases, an two groups, one concerned by mental occupational
explanation may be suggested in the light of Pearsall et al.’s stress including a significant physical effort and one
work (2009). They found out that challenge stressors for concerned by mental occupational stress only.
teams positively affected the performance. If we consider the - Each group of students could be divided in three
WPE cases as contexts with additional challenge stressors for sub-groups according to the extreme of the bell curve
associated to the Yerkes & Dodson curve and the
 Psychology and Social Behavior Research (2014) 20-28 27

HFS defined consequently. 60 yo.

- The consecutive finding was that most of the Further experiments are planned to analyze these points
students were not in good conditions of training and try to enlarge the application area of the protocol.
because of a too high level of stress.
- It suggested another approach of the training session.
For this last point, we can develop suggestions as follows, Declaration of conflict of interest
taking into account both our knowledge in the field of The authors declare no conflict of interest in relation to this
simulation training and the debriefing interviews of our article.
present experiments on simulators:
- Plan progressive simulation training.
- Each step of the progressive training is led by a Acknowledgments
specific pedagogical goal.
- The first step is a discovering period with no time The authors would like to thank Prof. D. Benhamou
pressure, no disturbance, and social support from the (Département d'Anesthésie-Réanimation de l'Hôpital de
physician-trainers. Bicêtre, Paris, France) and Prof. J.C. Granry (Pôle
- The sessions include unconscious integrating period Anesthésie-Réanimation, CHU d’Angers, France) for advice.
(nights).
It must be clear that, at this stage of the research, the
protocol includes performance and stress, not only stress. This
References
means that the protocol allows, for a sample of subjects, to Bailon, R., Mainardi, L., Orini, M., Sörnmo, L., & Laguna, P. (2010).
rate the individual performance and stress with regards to the Analysis of heart rate variability during exercise stress testing using
respiratory information. Biomedical Signal Processing and Control, 5,
group. It does not permit to rate the performance and stress of 299-310.
one subject alone, or to evaluate only subjects’ stress (without Beehr, T., & Newman, J. (1978). Job stress, employee health and
performance) inside the group. organizational effectiveness: a facet analysis, model, and literature
review, Personnel psychology, 31, 665-699.
The stress coefficient is fixed by the protocol while the
Bridger, R. S., Brasher, K., & Kilminster, S. A. D. (2011).. Job stressors in
performance coefficient determination depends on the work naval personnel serving on ships and in personnel serving ashore over a
activity and must be carefully designed and calculated. twelve month period. Applied Ergonomics, 42, 710-718.
These research results have thus limits depending on the Broadhurst, P. L. (1957). Emotionality and the Yerkes-Dodson law. Journal
of Experimental Psychology, 54, 345-352.
design of the study itself. These limits bound the application Brunet, A., Weiss, D., Metzler, T., Best, S., Neylan, Th., Rogers, C., Fagan, J.,
area of the protocol. & Marmar, Ch. (2001). The Peritraumatic Distress Inventory: A
One of the limits of the protocol developed here is a basic Proposed Measure of PTSD Criterion A2. Am J Psychiatry, 158(9),
1480-1485.
postulate designing the study: it concerns the type of stress. Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of
For long term stress, or for another kind of stress than perceived stress. Journal of Health and Social Behavior, 24, 385-396.
occupational stress, we cannot state whether our protocol may Davezies, Ph. (2008). Stress et pouvoir d’agir: données biologiques. Santé&
Travail, 64. Retrieved 01 March 2013 from
be applied or not. Theoretical considerations lead to the http://www.alternatives-economiques.fr/page.php?controller=article&action
assumption that the protocol cannot be applied to long term =htmlimpression&id_article=39496&id_parution=791
stress. De Jonckheere, J., Logier, R., Jounwaz, R., Vidal, R., & Jeanne, M. (2010).
From pain to stress evaluation using heart rate variability analysis:
At this stage of the research, the possibility to compare a development of an evaluation platform. Engineering in Medicine and
given sample of subjects in a given context to another one is Biology Society (EMBC), 2010 Annual International Conference of the
not studied. We just can claim that the protocol can describe IEEE, 3852-3855.
Drach-Zahary, A., & Erez, M. (2002). Challenge versus threat effects on the
with satisfactory accuracy the performance vs stress for a
goal–performance relationship. Organizational Behavior and Human
context of work. Nevertheless, it remains much less accurate Decision Processes, 88, 667-682.
than what could give the complex protocols including Fairclough, S., & Houston, K. (2004). A metabolic measure of mental effort.
analysis of the whole set of physiological parameters Biological Psychology, 66, 177-190.
Fauquet-Alekhine, Ph. (2012). Behavior as a consequence to fully describe
(Fairclough et al., 2004; Montano et al., 2009; Müller et al., short term occupational stress. Proceeding 17th Annual International
2009; Rohleder et al., 2009; Schubert et al. 2009; Bailon et al., "Stress and Behavior" Neuroscience and Biopsychiatry Conference, St.
2010; De Jonckheere et al., 2010; Jo et al., 2013). Yet it gives Petersburg, Russia, 31.
Fauquet-Alekhine, Ph., Frémaux, L., & Geeraerts, Th. (2011). Cognitive
a instant assessment of performance versus stress with a disorder and professional development by training: comparison of
satisfactory accurateness for training centers. simulator sessions for anaesthetists and for nuclear reactor pilots.
The simplifying hypothesis done to obtain the reduced Proceedings of the the XVe European Conf. on Developmental
Psychology, Pianoro (Italia): Medimond Srl., 83-87.
stress coefficient compared to the stress coefficient gives Fauquet-Alekhine, Ph., Geeraerts, Th., & Rouillac, L. (2012). Improving
another limit: results have been obtained with a sample of simulation training: anesthetists vs nuclear reactor pilots. In
subjects inside a narrow age interval; in this case, the Fauquet-Alekhine, Ph. (eds) Socio-Organizational Factors for Safe
Nuclear Operation, (pp. 32-44). Montagret: Larsen Science Ed., 1.
reference heart rate might have less influence in the Retrieved 01 March 2013 from http//hayka-kultura.com/larsen.html
calculation from one subject to another, which would perhaps Geeraerst, Th., Roulleau Ph., Cheisson, G., Aidan, K., Lallali, K., Leguen, M.,
not be the case within an interval including people of 25 and Schnell, D., Trabold, F., Fauquet-Alekhine, Ph., Duranteau, J., &
28 Psychology and Social Behavior Research (2014) 20-28

Benhamou, D. (2013). Physiological and psychological stress induced arterial blood pressure and anxiety-related behavior.
by a simulation course in anaesthesia residents (accepted in Simulation Psychoneuroendocrinology, 27, 549-561.
in Healthcare). Montano, N., Porta, A, Cogliati, Ch., Costantino, G., Tobaldini, E., Rabello
Hancock, P. A., Szalma, J. L., & Weaver, J. L. (2002). The distortion of Casali, K., & Iellamo, F. (2009). Heart rate variability explored in the
perceptual space-time understress. DoD Multidisciplinary Research frequency domain: A tool to investigate the link between heart and
Program: MURI Operator Performance Under Stress(OPUS), White behavior. Neuroscience and Biobehavioral Reviews, 33, 71-80.
Paper. Müller, M., Hänsel,M., Fichtner, A., Hardt, F., Weber, S., Kirschbaum, C.,
Jehel, L., Brunet, A., Paterniti, S., & Guelfi, J. (2005). Validation de la Rüder, S., Walcher, F., Koch, T., & Eich, Ch. (2009). Excellence in
version française de l’inventaire de détresse péritraumatique. Can J performance and stress reduction during two different full scale
Psychiatry, 50(1), 67-71. simulator training courses: A pilot study. Resuscitation. In Press.
Jehel, L., Paterniti, S., Brunet, A., Louville, P., & Guelfi, J. (2006). Palmer, S., Cooper, C., & Thomas, K. (2003). Revised model of
L’intensité de la détresse péritraumatique prédit la survenue des organisational stress for use within stress prevention/management and
symptômes post-traumatiques parmi des victimes d’agressions. wellbeing programmes – brief update. International Journal of Health
L’Encéphale, 1(32), 953-956. Promotion and Education. 41(2), 57-58.
Jo, N. Y., Lee, K. C., & Lee, D. S. (2013). Computer-Mediated Task Pearsall, M. J., Ellis, A. P. J., & Stein, J. H. (2009). Coping with challenge
Performance Under Stress and Non-stress Conditions: Emphasis on and hindrance stressors in teams: Behavioral, cognitive, and affective
Physiological Approaches. Digital Creativity, 32, 15-27. outcomes. Organizational Behavior and Human Decision Processes,
Judge, T. A., Ilies, R., & Zhang, Z. ( 2012). Genetic influences on core 109, 18-28.
self-evaluations, job satisfaction, and work stress: A behavioral Rohleder, N., & Nater, U. (2009). Determinants of salivary a-amylase in
genetics mediated model. Organizational Behavior and Human humans and methodological considerations.
Decision Processes, 117, 208-220. Psychoneuroendocrinology, 34, 469-485.
Karasek R. A., & Theorell R. (1990). Healthy work, Stress, Productivity and Schubert, C., Lambertz, M., Nelesen, R. A., Bardwell, W., Choi, J. B., &
the Reconstruction of Working Life, the Free Press. Basic books: New Dimsdale, J. E. (2009). Effects of stress on heart rate complexity—A
York. comparison between short-term and chronic stress. Biological
Keitel, A., Ringle, M., Schwartges, I., Weik, U., Picker, O., Stockhorst, U., & Psychology, 80, 325-332.
Deinzer, R. (2011). Endocrine and psychological stress responses in a Spielberger, C. D. (1983). Manual for the State-Trait Anxiety Inventory.
simulated emergency situation. Psychoneuroendocrinology, 36, Consulting Psychologist Press Inc., Palo Alto, California.
98-108. Staal, M. (2004). Stress, Cognition, and Human Performance: A Literature
Kilminster, S., & Bridger, R. S. (2007). Modified work and Well-Being Review and Conceptual Framework. NASA report, reference:
Questionnaire for Cohort Study of Stress in the Naval Service: NASA/TM—2004–212824.
Psychometric Validation. Unpublished MoD Report. Steptoe, A. (2000). Stress, social support and cardiovascular activity over the
Leonova, A. B. (2009). The concept of human functional state in Russian working day. International Journal of Psychophysiology, 37, 299-308.
applied psychology. Psychology in Russia: State of the Art, 517-538. Yerkes, R. M., & Dodson, J. D. (1908). The relation of strength of stimulus to
McLean, A. (1974). Concepts of occupational stress, in A. McLean (eds) rapidity of habit-formation. Journal of Comparative Neurology and
Occupational Stress, (pp. 3-14). Springfield, Illinois: Thomas. Psychology, 18, 459-482.
Maslova, L. N. , Bulygina,V. V. , & Markel, A. L. (2002). Chronic stress
during prepubertal development: immediate and long-lasting effects on