Document text
Report No. 06-09
Approved for public release, distribution unlimited..
Naval Health Research Center
140 Sylvester Road
San Diego , California 92106
Naval Health Research Center
Neurophysiologic Methods to Measure
Stress During Survival , Evasion ,
Resistance , and Escape Training
M. K. Taylor
K. P. Sausen, L.R. Mujica-Parodi
E. G Potterat, M. A. Yanagi
H. Kim
REVIEW ARTICLE
Neurophysiologic Methods to Measure Stress
During Survival, Evasion, Resistance, and Escape
Training
Marcus K. Taylor, Kenneth P. Sausen,
Lilianne R. Mujica-Parodi, Eric G. Potterat,
Matthew A. Yanagi, and Hyung Kim
Taylor MK, Sausen KP, Mujica-Parodi LR, Potterat EG,
Yanagi MA, Kim H. Neurophysiologic methods to measure stress
during survival , evasion , resistance, and escape training. Aviat Space
Environ Med 2007; 78(5, Suppl.):B224-30.
Training in Survival, Evasion, Resistance, and Escape (SERE) is re¬
quired for U.S. military members at high risk of capture. This physically
and psychologically demanding course is considered an analog to the
stress imposed by war, captivity, and related events, thus offering a
unique and unprecedented medium in which to systematically examine
human stress and performance during a realistically intense operational
context. Operational stress is multifaceted, manifesting cerebral, neu¬
roendocrine, cardiac, and cognitive characteristics, and necessitating an
integration of multiple methods of measurement to appropi-Wd^lYKar-
acterize its complexity. Herein we describe some of Dur^eseaLr^sea^h
methods and discuss their applicability to real-time monitoring and
predicting of key aspects of human performance. A systems approach is
taken, whereby some of the "key players" implicated imtjpe stress
response (e.g., cerebral, neuroendocrine, cardiac) areurtefly discuAft^
to which we link corresponding investigative techniques (fMRI, acoustic
startle eye-blink reflex, heart rate variability, and neuroendocrine sam¬
pling). Background and previous research with each investigative tech¬
nique and its relationship to the SERE context is briefly reviewed.
Ultimately, we discuss the operational applicability of each measure,
that is, how each may be integrated with technologies that allow com¬
putational systems to adapt to the performer during operational stress.
Keywords: cortisol, MRI, acoustic startle, heart rate variability.
TRAINING IN Survival, Evasion, Resistance, and
Escape (SERE), including a period of confinement
in a Resistance Training Laboratory (RTL) is required
for U.S. military members at high risk of capture. After
an initial phase of classroom-based didactic training,
students are taken to the field where they receive ap¬
plied training in survival, evasion, resistance, and es¬
cape techniques. Students are then released into the
field and tasked with the goal of evading enemy cap-
tors. On eventual capture, students are taken to the
RTL, where they are expected to apply their recently
learned skills of resistance to political indoctrination
and captivity-related problems. The structured, choreo¬
graphed nature of this training platform provides a
unique and unprecedented medium in which to sys¬
tematically and scientifically examine human stress and
performance in a realistic military context. Moreover,
since SERE training is designed to simulate the prison¬
er-of-war experience, it offers a unique medium in
which to study the effects of captivity stress on key
aspects of human performance. Our program of re¬
search is designed to identify the extent to which indi¬
vidual differences in brain and peripheral stress system
reactivity to (pre-SERE) laboratory tasks are predictive
of stress reactivity, military performance, and subse¬
quent memory for captivity-related events that occur
during training. Herein we review key methods we are
using to quantify individual differences in stress reac¬
tivity in an operationally relevant setting; describe how
thesp methods are being applied in a highly relevant
operational environment; and discuss possible opera¬
tional ^P^lfeability of each neurophysiologic measure
62lajive io real-time monitoring systems.
research methods presently being em¬
ployed in our study aimed at elucidating the multifac¬
eted nature of extreme operational stress are described
in this report. To begin, some of the "key players" of the
stress response (e.g., cerebral, neuroendocrine, cardiac,
and cognitive-behavioral systems) will be briefly dis¬
cussed, to which we will link specific investigative tech¬
niques used to characterize their respective responses
during SERE training (e.g., fMRI, acoustic startle eye-
blink reflex, heart rate variability, neuroendocrine sam¬
pling). Background and previous research with each
investigative technique in the SERE context will build
the "theoretical platform" for each method. Most im¬
portantly, we will discuss the operational applicability
of each neurophysiologic measure, how each may serve
From the Naval Health Research Center, San Diego, CA (M. K.
Taylor, K. P. Sausen, M. A. Yanagi); State University of New York,
Stony Brook, NY (L. R. Mujica-Parodi); and Fleet Aviation Specialized
Operational Training Group, Pacific, San Diego, CA (E. G. Potterat, H.
Kim).
Address reprint requests to: Marc Taylor, Ph.D., LT, MSC, USN,
Naval Health Research Center, PO Box 85122, San Diego, CA 92186-
5122; [email protected].
Operational Applications of Cognitive Performance Enhancement Tech¬
nologies was supported through the Office of Naval Research (ONR),
Arlington, VA; the U.S. Army Medical Research and Materiel Com¬
mand (USAMRMC), Ft. Detrick, MD; the Eye-Corn Eye-tracker Re¬
search Program, Reno, NV; and through an unrestricted educational
grant from Cephalon, Inc.
Reprint & Copyright © by Aerospace Medical Association, Alexan¬
dria, VA.
B224
Aviation , Space , and Environmental Medicine • Vol. 78, No. 5, Section II • May 2007
NEUROPHYSIOLOGIC METHODS & SERE— TAYLOR ET AL.
as a tool to monitor and predict stress reactivity, cog¬
nition, memory, and other key elements of human per¬
formance during operational stress. The Institutional
Review Board at the Naval Health Research Center has
approved this study, and each subject has provided
written informed consent prior to participating.
"Key Players" in the Stress Response
Cerebral Factors
The limbic system includes the hypothalamus, hip¬
pocampus, amygdala, and several related structures
(e.g., cingulate gyrus, septum, ventral tegmental area,
and prefrontal cortex). The amygdala is implicated in
emotional memory as well as the generation of anger
and fear, and is particularly important for modulating
the storage of emotionally charged memories (15). More
specifically, it has been shown that the amygdala is
particularly important for recalling the gist (though not
the details) of emotionally charged memories (2), and is
known to play a vital role in the memory of faces (1).
The hippocampus is primarily responsible for consoli¬
dation of declarative information into long-term mem¬
ory, and is particularly important for declarative, spa¬
tial, and contextual memory, and for modulation of the
hypothalamic-pituitary-adrenal (HPA) axis. The hip¬
pocampus and its surrounding structures, richly popu¬
lated with cortisol receptors, are particularly vulnerable
mined by the blood oxygen level-dependent signal. It is
expected that selectivity of amygdalar response to emo¬
tionally evocative vs. neutral faces will relate positively
to stress resilience during SERE. Also, in light of the
established link between amygdalar activation and
memory, it is predicted that increased amygdalar acti¬
vation in response to the faces task will be associated
with greater recall of SERE-related people and events.
We are also exploring associations between hippocam¬
pal and frontal activation, subsequent SERE perfor¬
mance, and memory of SERE events. Initial data from
our collaborative laboratory (Mujica-Parodi LR. Unpub¬
lished observations; June 1, 2006.) indicate specific pat¬
terns between amygdala reactivity and pre-pulse inhi¬
bition (PPI) of the startle reflex (discussed later), where
amygdala reactivity to threatening (vs. neutral) facial
stimuli is associated with reductions in PPI in response
to threatening stimuli. These data indicate that amyg¬
dala reactivity is implicated in resilience to mild labo¬
ratory stress, but more work is needed to establish the
predictability of amygdala reactivity relative to the ex¬
treme operational stress endured during SERE training.
These and subsequent SERE-specific data will be fully
addressed in separate reports. Finally, it is important to
note that our research distinguishes between general
limbic activation and fear- and anger-related activation.
Given the central role of the amygdala in the processing
of these emotions, future research should similarly ap-
to damage from glucocorticoids secreted during chronidw it
stress by the HPA axis, thus demonstrating a key neu- , , Operational impact: fMRI technology may be applica-
robiological link between stress and Memory. Stimula- ble^a ™eans profile individual differences in stress
tion of the hippocampus is known to result in hyper- an<^ Perf°rmance a priori. This information could then
jue 22 20tf7wover>1rlteemerSinS technologies to combine neu-
?t rophysiologic sensors and cognitive stc
vigilance and apprehension (3).
Background/precedence: The use of fMRI technology to
study cerebral responses in SERE-related research is
unprecedented. Our initial work (Mujica-Parodi LR, et
al. Unpublished observations; June 1, 2006) on individ¬
ual variability with respect to cognitive and physiolog¬
ical resilience to stress indicates that selectivity of
amygdalar response to threatening vs. nonthreatening
stimuli (rather than either the amplitude or habituation
of the amygdala response to threatening stimuli) is
predictive of physiological (cardiac, endocrine), cogni¬
tive, and self-perceived stress-resilience. Subjects
grouped by high and low selectivity show markedly
different patterns of neural activation; that is, individ¬
uals with high-selectivity demonstrate tight patterns of
activation centered exclusively on the limbic regions,
whereas individuals with low selectivity show diffused
and global cerebral activation.
Overview of current research strategy: In light of the
established links between key limbic structures, stress,
and memory, we are presently applying fMRI technol¬
ogy for noninvasive neuroimaging of the brain prior to
SERE training, and then exploring its ability to predict
key aspects of human performance and stress resilience
during training. Specifically, subjects are asked to view
a standard set of facial pictures developed by Eckman &
Friesen (8). Some of the faces display neutral expres¬
sions, while others show emotionally evocative faces
(e.g., anger, fear). Endpoints of interest include amyg¬
dalar, frontal, and hippocampal activation as deter¬
ophysiologic sensors and cognitive state gauges within
operational platforms, yielding human-computer sys¬
tems that are modifiable to individual neurophysiologic
input. More specifically, fMRI findings from this study
could provide additional information to pinpoint task-
specific cerebral regions of interest that may be affected
during intense military stress, as well as correlate the
environmental situations and patterns that elicit the
observed responses. This information could then be
integrated with emerging real-time field applications,
such as head-mounted neurophysiologic monitors de¬
signed to detect stress overload and mitigate stress-
induced performance decrement via reduction or redis¬
tribution of information or workload. Additionally,
once the environmental patterns and situations are
identified that elicit harmful responses, an inference
engine or classifier can be implemented to scan subse¬
quent incoming information and stimuli to "keep an
eye out" for future events of similar consequence. The
prerequisite, of course, is a clear understanding not
only of a causal link between the limbic structures of
interest and subsequent operational decrement, but also
of the particular patterns of brain activation in response
to operational stress, both of which this program of
research is designed to improve our understanding.
Although our protocol does not address this, it also
would be of special interest to examine functional brain
activation during and after SERE training. One obvious
limitation is that, although there are portable MRI ma-
Aviation, Space , and Environmental Medicine • Vol. 78 , No. 5, Section II • May 2007
B225
NEUROPHYSIOLOGIC METHODS & SERE— TAYLOR ET AL.
chines that could be used in the field, SERE students
would have to be taken out of the scenario for a sub¬
stantial period of time, thus detracting from their train¬
ing experience (Table I). Post-SERE FMRI, however,
would be an invaluable addition to the growing body of
knowledge. In sum, our findings are expected to eluci¬
date the brain regions implicated in the response to
extreme military stress, which will then aid in the iden¬
tification of regions of interest useful to emerging tech¬
nologies used to promote stress detection and mitigate
performance decrement in the operational domain. This
will be accomplished either via reduction, redistribu¬
tion, or modification of the performer's workload,
through moderation of environmental stimuli, or both.
Acoustic Startle Eye-Blink Response (ASER)
In addition to fMRI, another readily observable mea¬
sure with an established biological link to the limbic
system concerns the acoustic startle eye-blink response
(ASER). This measure, a useful probe for studying brain
activity linked with fear, anxiety, and other emotional
states in humans, is described in this section.
The ASER is elicited by an abrupt noise, which is
processed as a reflex from the auditory nerve by ventral
cochlear root neurons projecting to the nucleus reticu¬
laris pontis caudalis and on to the facial nerve inner¬
vating the orbicularis occuli (muscle that blinks the eye)
(6,12). The ASER is measured by placing surface elec¬
trodes beneath and beside the eye to record an mte-
grated electromyographic signal. The nucleus reticu-
Background/precedence: As is true with fMRI, the use of
ASER technology in the SERE context is unprecedented.
In the laboratory setting, however, Lang and colleagues
(5,10,11) have examined ASER in relation to visual stim¬
uli ranging from highly pleasant to highly unpleasant.
ASER was shown to increase in response to aversive
stimuli; likewise, it is reduced when the emotional con¬
tent of the picture is rated as pleasant or positive.
Overview of current research strategy: In our study, we
present to the participant several standardized photo¬
graphs from the International Affective Picture System
(IAPS), some of which are emotionally neutral, while
others are emotionally evocative. All emotionally neu¬
tral stimuli are presented on one day while evocative
stimuli are presented on the other day, and order of
presentation is counterbalanced. The task starts with a
2-min orienting cross followed by 48 IAPS pictures,
which are presented for 6 s each. During the task,
subjects hear a series of 10 "test" trials and 10 "control"
trials, interspersed in pseudorandom order with an in¬
terstimulus interval of 8-24 s. Test trials are composed
of a brief white-noise pulse (95 dB for 50 ms) preceded
by a white-noise prepulse (70 dB presented for 20 ms).
The beginning of the prepulse and the beginning of the
pulse stimulus are separated by 120 ms. Control trials
are composed of a pulse alone (95 dB for 50 ms, without
a prepulse). White noise pre-pulses have been shown to
exert an inhibitory effect on physiological responses
(i.e., sensory gating) to immediately subsequent stimuli
TubfiwTo non-prepulsed control stimuli.
laris pontis caudalis is innervated In lighUof Lang et al.'s findings (10,11) we hypothe-
amygdala, thus enabling it to modulate the ASER dur-8 [&M that participants will demonstrate an increased
ing emotion. In humans, the ampli tudupfj ^heoASER is Q^E^ae^anpf to aversive stimuli compared with the
increased by images that evoke fear or are otherwise
aversive (10). In essence, ASER is believed to be an
amygdala-modulated defensive reflex to aversive stim¬
uli when the motivational state is one of avoidance or
withdrawal. ASER technology can best be integrated
with other measures of stress reactivity, including gal¬
vanic skin responses (GSR) and heart rate variability
(HRV; discussed later), in order to appropriately char¬
acterize the complex, multidimensional nature of stress
reactivity. We have taken these steps in our present
research, integrating ASER with GSR, HRV, fMRI, and
neuroendocrine measures of stress.
neutral stimuli. We further expect that ASER ampli¬
tudes will relate to key endpoint measures of stress
reactivity during SERE training, including cortisol reac¬
tivity, psychological symptoms of dissociation, and de¬
clarative memory. Finally, we hypothesize that PPI of
the startle reflex will be reduced during the emotionally
evocative condition, and will inversely relate to stress
reactivity during SERE training as measured by sali¬
vary cortisol reactivity. Our pilot work (Mujica-Parodi
et al., unpublished observations; June 1, 2006) indicate
less pre-pulse inhibition of the startle reflex in the pres¬
ence of threatening (vs. neutral) stimuli. These data and
TABLE I. SUMMARY OF EXPERIMENTAL TECHNIQUES USED IN SERE RESEARCH.
Technique
Key Measurement Factor
Advantages
Disadvantages
fMRI
Amygdala function. Hippocampal
structure and function, and
frontal activation
Specifies neuroanatomical regions of interest
May be impractical for real-time
monitoring
ASER
Startle reflex; amygdala-
modulated defensive reflex to
aversive stimuli
Accessible, portable, affordable, minimally
invasive; easily synchronized with other
stress measures (HRV, GSR) with
available software
Possible technology limitations
HRV
Sympathetic and parasympathetic
activation
Sensitive to changes in emotional state,
stress, and physical exertion; easily
synchronized with other measures
Lack of real-time measurement
capabilities
N euroendocrine
Sampling
HP A/ Sympathomedullary
activation
Strong scientific basis linking to stress
(predictive power); portable detection
systems and patch technology may be
available options for operational use
No apparent disadvantages
B226
Aviation , Space , and Environmental Medicine • Vol. 78, No. 5, Section II • May 2007
NEUROPHYSIOLOGIC METHODS & SERE— TAYLOR ET AL.
subsequent links between ASER, PPI, and SERE perfor¬
mance will be addressed in separate reports.
Operational impact: ASER technology is fairly accessi¬
ble, portable, and reasonably affordable. We anticipate
that this technology will be useful as a means to profile
individual differences in stress and performance a pri¬
ori, and it is also conceivable that ASER technology
could be integrated into closed-loop human computer
interface (HCI) systems (e.g., head-mounted neuro¬
physiologic monitors) designed to improve cognitive
performance under operational stress (Table I). Specif¬
ically, ASER can help to identify acute threat responses
(specific time points when the performer is responding
to a perceived environmental threat). HCI systems
could then accommodate these momentary deviations
in affective state of the performer by redistributing or
reducing workload, deleting the stressor from the per¬
former's sensory field, and/or by implementing strate¬
gies for stress mitigation (e.g., relaxing music). By ex¬
tension, since prepulses have been shown to inhibit the
startle reflex, it is also possible that if monitoring sys¬
tems could anticipate (through threat recognition) the
environmental stressor causing the performer to startle,
it could systematically attach prepulse stimuli to inhibit
startle. These strategies would help to filter the threat¬
ening stimulus from the performer's sensory field or
mitigate its effects, thereby promoting attentional con¬
trol and enhancing operational performance. It is also
possible that ASER amplitudes could be converted into
discussed here, along with a description of its use in our
study as well as its possible operational impact.
HRV refers to the variation in heart rate between
successive beats, and is commonly described by the
standard deviation of intervals between successive R
waves in the cardiac cycle. Short-term variation can be
decomposed mathematically into components of the
frequency spectrum that estimate autonomic modula¬
tion of heart rate. In particular, the high frequency
component (0.15-0.5 Hz) is believed to correspond to
modulation of the cardiac cycle by vagus nerve stimu¬
lation, reflecting rapid transmission of acetylcholine in
inhibiting heart cells by directly opening ion chan¬
nels. Conversely, the low-frequency (LF) spectrum
(0.05-0.15 Hz) corresponds to baroreflex control of
heart rate and reflects mixed sympathetic and parasym¬
pathetic modulation. This response is slower, primarily
because the action of norepinephrine on heart cells
depends on a second-messenger system to open ion
channels, in contrast to the direct action taken by ace¬
tylcholine. HRV, then, is a noninvasive test of auto¬
nomic function which can be dramatically affected by
emotional states, stress, and physical exertion (9). As
alluded to earlier relative to ASER technology, HRV
should be integrated with other measures of stress re¬
activity to more fully appreciate the complexity under¬
lying stress reactivity — a step we have taken in our
present research.
Background/precedence: Anxiety and stress have been
gauges made accessible directly to the performer re- u9Sfilat&d \\dth lowered HRV, causing a reduction in
garding his or her startle responses ifidte operatiMd parasymp^theiidl(vagal) activity and corresponding in
environment. This form of biofeedback could beiused as*
a training apparatus to aid the performer in making
cognitive or tactical adjustments to adapt to the stressor
and, subsequently, improve his or her capacity to cope
with environmental demands and prevent performance
decrement. Ultimately, it is expected that ASER meth¬
odologies could be best utilized by integrating with
other established methods of monitoring, including gal¬
vanic skin response (GSR) and heart rate variability
(HRV), to better characterize and accommodate for
acute stress responses in the performer. Such triangu¬
lation of data is likely to sense the performer's state
with greater accuracy, because multiple methods of
measurement converging at a single time point provide
greater validity that any single measure alone. To this
end, we are employing complex systems analysis, an
analytic approach involving the application of complex
mathematical models to demonstrate how multiple
stress systems behave in synchrony.
In sum, ASER may be a useful instrument not only as
a predictor of stress response to environmental threats,
but also as a tool to perform real-time monitoring of a
performer's perceived threat in the operational domain.
Heart Rate Variability
While ASER is an observable measure of avoidance
and withdrawal that is biologically linked to the limbic
system, HRV refers to an observable measure of auto¬
nomic function that is sensitive to changes in emotional
states, stress, and physical exertion. This measure is
f$®ease in sympathetic tone. Pagani et al. (24) related
^Ijnpss in Japu^ans to an increase in the low frequency
variability (LF: 0.04-0.15 Hz) of heart rate, and Dinca-
Panaitescu et al. (7) and others (25) found that increases
in the QT interval LF component are more specific to
mental stress. Additionally, evidence suggests that cir¬
cadian rhythm of HRV is blunted in rats during chronic
stress (31), a finding that has also been reported in
humans with coronary artery disease. HRV has also
been shown to correlate with cortisol and norepineph¬
rine elevation (13), as well as lowered immune response
after stress exposure (4). Furthermore, data from our
collaborative laboratory (Mujica-Parodi LR, et al. Un¬
published observations; June 1, 2006.) demonstrate
links between HRV and amygdala function as well as
perceived stress, trait anxiety, and trait anger.
Overview of current research strategy: HRV analysis
represents another novel research method being ap¬
plied in SERE research. In the present study, 24-h elec¬
trocardiographic recordings are performed prior to
SERE training using the Aria Holter monitor (Del Mar
Reynolds Medical, Irvine, CA) with a sampling rate of
128 samples per second. The records are then reviewed
and edited using the Impresario Holter Analysis system
(Del Mar Reynolds Medical, Irvine, CA). In correla¬
tional analyses, HRV during a 24-h period is expected
to relate to perceived stress, trait anxiety, and trait
anger. In prospective analyses, HRV is expected to pre¬
dict cortisol reactivity, dissociation, and declarative
memory during SERE training.
Operational impact: This program of research will help
Aviation, Space, and Environmental Medicine • Vol. 78, No. 5, Section II • May 2007
B227
NEUROPHYSIOLOGIC METHODS & SERE— TAYLOR ET AL.
to determine whether HRV may be useful as a means to
profile individual differences in response to extreme
military stress. It appears that HRV could be integrated
into augmented cognition platforms and related tech¬
nologies designed to provide cognitive feedback to an
HCI system. As discussed earlier, it is expected that the
integration of multiple methods to detect stress re¬
sponses (e.g., HRV, GSR, and ASER) is needed to opti¬
mally monitor, appropriately characterize, and effec¬
tively accommodate acute stress responses in the
performer. A central limitation, however, concerns the
ease with which HRV data could be processed in real¬
time (Table I). At present, data analytic methods for
HRV are tedious and time consuming, and advance¬
ment in these methods is a crucial prerequisite to the
successful application of HRV technology during
stressful military operations.
Neuroendocrine Sampling
As ASER and HRV refer to observable measures of
biological processes implicated in operational stress,
HPA axis reactivity is also observable through nonin-
vasive sampling of stress hormones. Our method for
quantifying HPA reactivity to SERE training is dis¬
cussed in this section, along with a description of its use
in our work as well as its future operational impact.
Perhaps the most well known physiological response
to stress concerns the HPA axis. Emotional respqnse§ Joi
pal neurons (27-29), and may further suppress normally
occurring neurogenesis.
Unlike the research methods discussed earlier, mea¬
surement of cortisol responses and other stress-related
hormones is not without precedence in SERE research
(17-21). Morgan and associates (19), for instance, col¬
lected salivary data from 109 Army SERE students at
baseline, during 4 stress exposure time points during
SERE training, and at recovery. Cortisol increased sig¬
nificantly during the captivity experience, was greatest
after exposure to captivity-related problems, and re¬
mained elevated during recovery. These researchers
have also studied neuropeptide-Y (NPY) at baseline,
during a captivity phase of SERE training, and after
conclusion of training (21). In this study of 49 Army
SERE students, NPY levels were significantly elevated
compared with baseline following captivity-related
problems and were significantly higher in Special
Forces soldiers, compared with their non-Special Forces
counterparts. NPY was positively related to both corti¬
sol and behavioral performance under stress, and in¬
versely related to psychological symptoms of dissocia¬
tion, thus implying a stress-buffering effect of NPY.
These findings were replicated in a similar study con¬
ducted at the Navy SERE School (18). In yet another
study of neuroendocrine reactivity to SERE stress, Mor¬
gan et al. (20) demonstrated that cortisol release ac¬
counted for significant portions of variance in psycho¬
logical symptoms of dissociation (22%) as well as
perceived threat are generated in the limbic'sysfem, and military performance (31%). Finally, these scientists (22)
signals are sent to subcortical brainx^Eff^^fiSl the hav4 ShfMfe rebbhtly demonstrated that dehydroepi-
hypothalamus. The hypothalamus, in turn, activates i&8 L (^droiterone sulfate (DHEAS)/ cortisol ratios were sig-
posterior pituitary to secrete vasopressirTgpd Q^J^cantlyj higher in subjects who reported fewer symp
and the adrenal medulla to secrete epinephrine and
norepinephrine. Similarly, the anterior pituitary re¬
leases adrenocorticotropic hormone, which activates
the adrenal cortex to release glucocorticoids. The pri¬
mary glucocorticoid is cortisol, which mobilizes energy
for action and inhibits the immune response (23).
Background/precedence: Captivity is known to elicit el¬
evated levels of cortisol. For example, free-ranging
monkeys that were subsequently captured and held for
45 d manifested significantly higher cortisol levels dur¬
ing captivity than free-ranging monkeys who were
killed by hunters just prior to blood sampling (30). The
later group presumably had no substantial systematic
threat prior to being killed. Restraint stress, an animal
model of captivity, consistently results in elevated
stress-related hormones and neurotransmitters (16).
SERE training has likewise been shown to result in
significantly elevated catecholamine levels (20).
Excessive cortisol levels resulting from chronic expo¬
sure to severe stress (as occurs during SERE training)
are also known to degrade recall (14,26,29). Specifically,
cortisol evokes suppressing effects on the hippocampus
(14) and degrades perceptual processes important not
only for encoding information (i.e., tunnel vision) but
also for memory of salient information (29). Acute ex¬
posure to high cortisol levels results in degraded de¬
clarative memory, which is reversible on return to nor¬
mal levels. By contrast, chronic exposure to elevated
levels of cortisol has been shown to damage hippocam-
toms of dissociation and exhibited superior military
performance during SERE training, further implying a
stress-buffering role of DHEAS.
Overview of current research strategy: In our research,
an extensive panel of pre-SERE cortisol samples are
taken at regularly scheduled times throughout a 2-d
period in order to establish a normal baseline of diurnal
patterns. Cortisol responses to a series of mild cognitive
laboratory stressors are also measured prior to initia¬
tion of SERE training. Next, we assess cortisol reactivity
at several time points during the captivity phase of
SERE training, including a particularly challenging
training event. In total, we establish baseline cortisol
measures with extensive pre-SERE sampling, to which
we compare cortisol responses to acute military stress
(i.e., postevasion SERE captivity). It is expected that
baseline cortisol values as well as cortisol reactivity to
mild laboratory stressors will predict cortisol reactivity
during SERE training. Also, in agreement with Morgan
et al/s findings, our current data demonstrate that cor¬
tisol during captivity differs rather dramatically from
baseline, and peaks directly after the challenging train¬
ing events. These findings are in preparation for report
in subsequent publications. We are further exploring
individual differences in cortisol reactivity to SERE
training, as well as its relation to other neuroendocrine
responses.
Operational impact: The next step is to explore the
applicability of the present neuroendocrine research to
B228
Aviation , Space , and Environmental Medicine • Vol. 78, No. 5, Section II • May 2007
NEUROPHYSIOLOGIC METHODS & SERE— TAYLOR ET AL.
mitigate performance decrements in the performer or
warfighter. A possible application can be found in "skin
patches" currently being researched to monitor bio¬
chemical responses to operational stress via detection of
cortisol, catecholamines, DHEA, NPY, other relevant
stress hormones, alarm pheromones, and related chem¬
icals (Table I). Detection of extreme amounts of cortisol
via patch technology, for instance, could be fed back to
a monitoring system that could then make adjustments
in the workload or complexity of tasking for the over¬
stressed performer, thus bringing his or her cortisol
levels back within a predetermined effective range. Al¬
ternatively, a feedback system could be designed,
where the patch is programmed to counter the detected
stress responses with input of naturally occurring
stress-countering substances such as DHEA or NPY, or
a pharmacologically based alternative.
Summary
In the preceding discussion, we described several
research methods aimed at capturing the multifaceted
nature of acute operational stress during SERE training;
we briefly reviewed some of the key players implicated
in the stress response, to which we linked specific in¬
vestigative techniques (e.g., fMRI, acoustic startle reflex,
HRV, and neuroendocrine analyses) presently em¬
ployed in the SERE study; we discussed the operational
relevance of each neurophysiologic measure, how each
ernment. Approved for public release; distribution is unlimited. This
research has been conducted in compliance with all applicable federal
regulations governing the protection of human subjects in research.
REFERENCES
1. Adolphs R, Tranel D. Amygdala damage impairs emotion recog¬
nition from scenes only when they contain facial expressions.
Neuropsychologia 2003; 41(10):1281-9.
2. Adolphs R, Tranel D, Buchanan TW. Amygdala damage impairs
emotional memory for gist but not details of complex stimuli.
Nat Neurosci 2005; 8:512-8.
3. Buckworth J, Dishman RK. Exercise psychology. Champaign, IL:
Human Kinetics Publishers; 2002.
4. Cacioppo JT, Malarkey WB, Kiecolt-Glaser JK, et al. Heterogene¬
ity in neuroendocrine and immune responses to brief psycho¬
logical stressors as a function of autonomic cardiac activation.
Psychosom Med 1995; 57:154-64.
5. Cuthbert BN, Bradley MM, Lang PJ. Probing picture perception:
activation and emotion. Psychophysiology 1996; 33:103-11.
6. Davis M. The neurophysiological basis of acoustic startle modu¬
lation: research on fear motivation and sensory gating. In: Lang
PJ, Simons RF, Balaban M, eds. Sensory and motivational
processes. Mahwah, NJ: Erlbaum; 1997.
7. Dinca-Panaitescu S, Dinca-Panaitescu M, Achim A, Negoescu R.
Idioventricular low frequency oscillation in QT interval re¬
sponds univocally to RR confusing kinds of mental stress.
Integr Physiol Behav Sci 1999; 34:10-8.
8. Eckman P, Friesen WV. Pictures of facial affects. Palo Alto, CA:
Consulting Psychologists Press; 1975.
9. ESOC. Heart rate variability: Standards of measurement, physio¬
logical interpretation and clinical use. Circulation 1996; 93:
1043-65.
10. Lang PJ. The emotion probe. Studies of motivation and attention.
, , , j. . j , , , Am Psychol 1995; 50:372-85.
may serve as a tool to predict and momtMB^fga^y ImgQlgajJCtedley MM, Cuthbert BN. Emotion, motivation, and
tivity, cognition, memory, and other key elements qf . am^ty:T>rain mechanisms and psychophysiology. Biol Psychi-
l - - f - — -I™* - IVlCUlCdl ^^yi9W44fl248-63.
6iL&.eb2Lopez DE, Meloni EG, Davis M. A primary acoustic startle
pathiviayi^obligatory role of cochlear root neurons and the
mucteus reticularis pontis caudalis. J Neurosci 1996; 16:3775-
89.
13. Lovallo WR, Pincomb GA, Brackett DJ, Wilson MF. Heart rate
reactivity as a predictor of neuroendocrine responses to aver¬
sive and appetitive challenges. Psychosom Med 1990; 52:17-26.
14. McEwen BS. Plasticity of the hippocampus: adaptation to chronic
stress and allostatic load. Ann N Y Acad Sci 2001; 933:265-77.
15. McGaugh JL. The amygdala modulates the consolidation of mem¬
ories of emotionally arousing experiences. Ann Rev Neurosci
2004; 27:1-28.
16. Minton JE. Function of the hypothalamic-pituitary-adrenal axis
and the sympathetic nervous system in models of acute stress
in domestic farm animals. J Anim Sci 1994; 72:1891-8.
17. Morgan CA, Hazlett G, Wang S, et al. Symptoms of dissociation
in humans experiencing acute, uncontrollable stress: a prospec¬
tive investigation. Am J Psychiatry 2001; 158:1239-47.
18. Morgan CA, Rasmusson AM, Wang S, et al. Neuropeptide- Y,
cortisol, and subjective distress in humans exposed to acute
stress: replication and extension of previous report. Biol Psy¬
chiatry 2002; 52:136-42.
19. Morgan CA, Wang S, Mason J, et al. Hormone profiles in humans
experiencing military survival training. Biol Psychiatry 2000;
47:891-901.
20. Morgan CA, Wang S, Rasmusson A, et al. Relationship among
plasma cortisol, catecholamines, neuropeptide Y, and human
performance during exposure to uncontrollable stress. Psycho¬
som Med 2001; 63:412-22.
21. Morgan CA, Wang S, South wick SM, et al. Plasma neuropep-
tide-Y concentrations in humans exposed to military survival
training. Biol Psychiatry 2000; 47:902-9.
22. Morgan CA, Southwick S, Hazlett G, et al. Relationships among
plasma dehydroepiandrosterone and cortisol levels, symptoms
of dissociation and objective performance in humans exposed
to acute stress. Arch Gen Psychiatry 2004; 61:819-25.
23. Nemeroff CB. The neurobiology of depression. Sci Am 1998;
278:42-9.
human performance during operational stress.
By no means are these methods all-inclusive
proaches to characterizing human stress odiMhtvfec
treme environments and associated performance decre¬
ment. It may be beneficial, for instance, to combine
spatial measures (e.g., fMRI) with more temporal meth¬
ods (e.g., EEG, ERP) which may give more insight into
how quickly someone may or may not react to certain
stressors. Such integration of methodologies may be
beneficial for identifying predictors of resilience a pri¬
ori. Additionally, it is critical to appreciate that stress
responses and resultant human performance outcomes
is an inherently complex phenomenon, and the use of
multiple methods to characteristic physiologically dis¬
tinct (but possibly cross-correlated) mechanisms is
greatly preferred over any single measure alone. Fi¬
nally, the realistically stressful SERE environment con¬
fers unprecedented ecological validity. This, in turn,
may lead to new knowledge and associated neurophys¬
iologic monitoring techniques leading to new treatment
options for combat stress and post-traumatic stress dis¬
order.
ACKNOWLEDGMENTS
This study is supported by the Office of Naval Research, Award
No. N0001406WX20141. Appreciation is extended to Michelle Stoia
and Sue Sobanski for technical assistance. We also would like to
express appreciation to the members of HS-10 Helicopter Antisubma¬
rine Squadron, San Diego, CA, and Naval Special Warfare Center,
Coronado, CA for their participation in and support of this study. The
views expressed in this article are those of the authors and do not
reflect the official policy or position of the Department of the Navy,
Department of the Army, Department of Defense, or the U.S. Gov-
Aviation, Space , and Environmental Medicine • Vol. 78 , No. 5, Section II • May 2007
B229
NEUROPHYSIOLOGIC METHODS & SERE— TAYLOR ET AL.
24. Pagani M, Furlan R, Pizzinelli P, et al. Spectral analysis of R-R and
arterial pressure variabilities to assess sympatho-vagal interaction
during mental stress in humans. J Hypertens 1989; 7(6, Suppl):S14-5.
25. Piccirillo G, Viola E, Bucca C, et al. QT interval dispersion and
autonomic modulation in subjects with anxiety. J Lab Clin Med
1999; 133:461-8.
26. Roozendaal B. Systems mediating acute glucocorticoid effects on
memory consolidation and retrieval. Prog Neuropsychophar-
macol Biol Psychiatry 2003; 27:1213-23.
27. Sapolsky RM. Stress, glucocorticoids, and damage to the nervous
system: the current state of confusion. Stress 1996; 1:1-19.
28. Sapolsky RM. Glucocorticoids and hippocampal atrophy in neu¬
ropsychiatric disorders. Arch Gen Psychiatry 2000; 57:925-35.
29. Sapolsky RM. Stress and plasticity in the limbic system. Neuro-
chem Res 2003; 28:1735-42.
30. Suleman MA, Wango E, Sapolsky RM, et al. Physiologic manifes¬
tations of stress from capture and restraint of free-ranging male
African green monkeys (Cercopithecus aethiops). J Zoo Wildl
Med 2004; 35:20-4.
31. Takeuchi H, Enzo A, Minamitani H. Circadian rhythm changes in
heart rate variability during chronic sound stress. Med Biol
Eng Comput 2001; 3 9:113-7.
Delivered by Ingenta to:
Aerospace Medical Association
IP: 138.162.5.12
Tue, 22 May 2007 18:53:02
B230
Aviation , Space , and Environmental Medicine • Vol. 78, No. 5, Section II • May 2007
REPORT DOCUMENTATION PAGE
The public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data
sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other
aspect of this collection of information, including suggestions for reducing the burden, to Washington Headquarters Services, Directorate for Information Operations and
Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302, Respondents should be aware that notwithstanding any other provision of law, no person shall
be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB Control number. PLEASE DO NOT RETURN YOUR
FORM TO THE ABOVE ADDRESS.
12 DISTRIBUTION/AVAILABILITY STATEMENT
Approved for public release; distribution is unlimited.
13. SUPPLEMENTARY NOTES
Published in: Aviation. Space, and Environmental Medicine. 2007, 78(5, Suppl.) B224-230)
14. ABSTRACT (maximum 200 words)
Introduction: U.S. military members at high risk of capture are required to attend Survival, Evasion, Resistance, and
Escape (SERE) training. This physically and psychologically demanding course of training is considered a reasonable
analogue to the stress imposed by war, captivity, and related catastrophic events. SERE training offers the scientist a
unique and unprecedented medium in which to systematically examine human stress and performance during a
controlled, yet realistically intense, operational context. Operational stress is multifaceted — manifesting cerebral,
neuroendocrine, cardiac, and cognitive-behavioral characteristics — necessitating an integration of multiple methods of
measurement to appropriately characterize its complexity. Methods: This paper describes some of our present research
methods with potential for applicability to monitoring and predicting key aspects of human performance in operational
contexts. To begin, a systems approach is taken, whereby some of the “key players” implicated in the stress response
(e.g., cerebral, neuroendocrine, and cardiac systems) is briefly discussed, to which we link corresponding investigative
techniques (e.g., FMRI, acoustic startle eyeblink reflex, heart rate variability, neuroendocrine sampling) presently
employed in the SERE study. Background and previous research with each investigative technique in the SERE context is
briefly reviewed. Operational Relevance: Ultimately, we discuss the operational applicability of each neurophysiologic
measure, how each may be integrated with personal monitoring systems designed to enhance performance during
operational stress.
15. SUBJECT TERMS
cortisol, MRI, acoustic startle, heart rate variability _
17. LIMITATION 1 18. NUMBER
OF ABSTRACT OF PAGES
UNCL 7
1 19b. TELEPHONE NUMBER (INCLUDING AREA CODE)
Standard Form 298 (Rev. 8-98)
Prescribed by ANSI Std. Z39-18
19a. NAME OF RESPONSIBLE PERSON
Commanding Officer
16. SECURITY CLASSIFICATION OF:
a. REPORT I b. ABSTRACT I c. THIS PAGE
UNCL I UNCL I UNCL
1. Report Date (DD MM YY) 2. Report Type
06 02 06 Research
4. TITLE AND SUBTITLE
Neurophysiologic Methods to Measure Stress During Survival, Evasion,
Resistance, and Escape Training
6. AUTHORS
M. K. Taylor, K. P. Sausen, L. R. Mujica-Parodi, E. G. Potterat, M. A.
Yanagi & H. Kim
7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES)
Naval Health Research Center
P.O. Box 85122
San Diego, CA 92186-5122
8. SPONSORING/MONITORING AGENCY NAMES(S) AND ADDRESS(ES)
Chief, Bureau of Medicine and Surgery
Code M53
2300 E St NW
Washington DC 20372-5300
3. DATES COVERED (from - to)
AUG 06-FEB 06
5a. Contract Number:
5b. Grant Number:
5c. Program Element:
5d. Project Number:
5e. Task Number:
5f. Work Unit Number: 60524
9. PERFORMING ORGANIZATION REPORT
NUMBER
Report No. 06-09
10. Sponsor/Monitor's Acronyms(s)
11. Sponsor/Monitor's Report Number(s)