Activation of the arousal response can impair performance on a

Document Sample
Activation of the arousal response can impair performance on a Powered By Docstoc
					Activation of the arousal response can impair
performance on a simple motor task
J. Timothy Noteboom, Monika Fleshner and Roger M. Enoka
J Appl Physiol 91:821-831, 2001.

You might find this additional info useful...

This article cites 39 articles, 5 of which can be accessed free at:

This article has been cited by 5 other HighWire hosted articles
          Sex differences in response to cognitive stress during a fatiguing contraction
          Tejin Yoon, Manda L. Keller, Bonnie Schlinder De-Lap, April Harkins, Romuald Lepers and
          Sandra K. Hunter
          J Appl Physiol, November , 2009; 107 (5): 1486-1496.
          [Abstract] [Full Text] [PDF]

           Get Aroused and Be Stronger: Emotional Facilitation of Physical Effort in the Human
           Liane Schmidt, Marie-Laure Cléry-Melin, Gilles Lafargue, Romain Valabrègue, Philippe
           Fossati, Bruno Dubois and Mathias Pessiglione

                                                                                                                                     Downloaded from on May 7, 2011
           J. Neurosci., July 29, 2009; 29 (30): 9450-9457.
           [Abstract] [Full Text] [PDF]

           Sympathetic outflow enhances the stretch reflex response in the relaxed soleus muscle in
           Nis Hjortskov, Jørgen Skotte, Christian Hye-Knudsen and Nils Fallentin
           J Appl Physiol, April 1, 2005; 98 (4): 1366-1370.
           [Abstract] [Full Text] [PDF]

           Declarative Memory after Stress in Humans: Differential Involvement of the β-Adrenergic
           and Corticosteroid Systems
           Françoise S. Maheu, Ridha Joober and Sonia J. Lupien
           JCEM, March 1, 2005; 90 (3): 1697-1704.
           [Abstract] [Full Text] [PDF]

           The 1- to 2-Hz oscillations in muscle force are exacerbated by stress, especially in older
           Evangelos A. Christou, Jennifer M. Jakobi, Ashley Critchlow, Monika Fleshner and Roger M.
           J Appl Physiol, July 1, 2004; 97 (1): 225-235.
           [Abstract] [Full Text] [PDF]

Updated information and services including high resolution figures, can be found at:

Additional material and information about Journal of Applied Physiology can be found at:

This infomation is current as of May 7, 2011.

Journal of Applied Physiology publishes original papers that deal with diverse areas of research in applied physiology, especially
those papers emphasizing adaptive and integrative mechanisms. It is published 12 times a year (monthly) by the American
Physiological Society, 9650 Rockville Pike, Bethesda MD 20814-3991. Copyright © 2001 by the American Physiological Society.
ISSN: 0363-6143, ESSN: 1522-1563. Visit our website at
                                                                                                                            J Appl Physiol
                                                                                                                        91: 821–831, 2001.

Activation of the arousal response can impair
performance on a simple motor task

            Department of Kinesiology and Applied Physiology,
            University of Colorado, Boulder, Colorado 80309-0354
            Received 2 November 2000; accepted in final form 23 March 2001

   Noteboom, J. Timothy, Monika Fleshner, and Roger                  performance (10, 21, 27). Because of this interaction,
M. Enoka. Activation of the arousal response can impair              elevated neuroendocrine activity during heightened
performance on a simple motor task. J Appl Physiol 91:               arousal has been implicated as a factor that can alter
821–831, 2001.—The purpose of this study was to determine            motor output (22, 23).
the effect of arousal in men and women on the moment-to-                One theory, known as the inverted-U hypothesis
moment performance of a simple motor task. We examined
                                                                     (46), suggests that performance is maximized at inter-
the control of a precision task in the presence and absence of
imposed stressors. Twenty-nine subjects (14 men, 15 women;
                                                                     mediate levels of arousal (28, 30, 38, 45). For example,
18–44 yr) were randomly assigned to either a control group           scores in rifle shooting (40) and performance on a

                                                                                                                                               Downloaded from on May 7, 2011
or one of two stressor groups, Mental Math or Electric Shock.        throwing task (45) were found to vary with the level of
Subjects presented with Math and Shock stressors, which              anxiety. Such studies, however, have not examined the
lasted 10 min, experienced significant increases in cognitive         variation in arousal and performance during the exe-
and physiological arousal compared with baseline and con-            cution of a task and usually did not consider possible
trol subjects. Heart rate, systolic blood pressure, and electro-     sex differences in arousal. For example, female rats
dermal activity were elevated 5–80% with presentation of             exposed to various stressors, including electric shock,
the stressors, whereas diastolic blood pressure and salivary         alcohol, and immobilization, responded with elevated
cortisol were unchanged. The greater levels of cognitive and         levels of several neuroendocrine substrates compared
physiological arousal were associated with reductions in             with male rats (20, 34, 35). Similarly, psychologically
steadiness of a pinch grip for the Shock subjects ( 130%             demanding tasks evoked elevated cardiovascular ad-
reduction from baseline) but not for the subjects in the Math
                                                                     justments, immune system substrates, and neuroendo-
group, who experienced heightened arousal but no change in
steadiness (10% reduction from baseline). Although women
                                                                     crine responses in women compared with men (25, 29).
exhibited more of a reduction in steadiness than men, the            These differences in the arousal response are likely
effect was largely unrelated to the magnitude of the change          associated with sex-specific variations in the level of
in arousal.                                                          neuromodulatory agents in the central nervous sys-
stressor; steadiness; pinch task; force
                                                                        To investigate the association between arousal and
                                                                     motor output, we examined the ability of men and
                                                                     women to perform a precision task in the presence
AROUSAL IS A COMPONENT    of several emotional responses,            and absence of imposed stressors. Subjects were exposed
including anxiety and fear, and is characterized by                  to one of two laboratory stressors, mental math or electric
feelings of apprehension, nervousness, and tension.                  shock. The purpose was to determine the effect of
The physiological manifestations of arousal include                  arousal in men and women on the moment-to-moment
increased blood pressure, heart rate, sweating, dryness              performance of a simple motor task. We expected to
of the mouth, hyperventilation, and musculoskeletal                  find that the steadiness of a submaximal pinch task
disturbances, such as restlessness, tremors, and feel-               would be reduced during the stressor conditions and
ings of weakness (2, 3, 18, 30, 44). Although emotions               that the effect would be greater in women. Preliminary
can be perceived in the cerebral cortex, the predomi-                results have been presented in abstract form (33).
nant outcome is the automatic effects mediated by the
autonomic, endocrine, and neuromuscular systems.                     METHODS
The automatic responses involve subcortical parts of                    Twenty-nine subjects (14 men, 15 women; 18–44 yr) were
the nervous system, especially connections between                   tested in this study. The subjects were recruited from the
the nuclei of the amygdala, hypothalamus, and brain                  general population. The subjects had no history of mental
stem (19). The neurotransmitters and hormones that                   pathology or neuromuscular disease, including any diagnosis
elicit these responses, however, can potentially modu-               of an anxiety disorder or upper extremity injury for at least 6
late the function of the spinal circuits underlying motor
                                                                       The costs of publication of this article were defrayed in part by the
  Address for reprint requests and other correspondence: R. M.       payment of page charges. The article must therefore be hereby
Enoka, Dept. of Kinesiology and Applied Physiology, Univ. of Colo-   marked ‘‘advertisement’’ in accordance with 18 U.S.C. Section 1734
rado, Boulder, CO 80309-0354 (E-mail:     solely to indicate this fact.             8750-7587/01 $5.00 Copyright © 2001 the American Physiological Society                                   821
822                                         AROUSAL AND MOTOR PERFORMANCE

mo. Subjects were randomly assigned to one of three groups:          was assessed at multiple intervals during the experimental
a control group (5 women, 4 men; mean SE age: 28.2 2.7               session. The experiments were conducted with the subjects
yr; range: 18–45 yr), a Mental Math group (5 women, 5 men;           seated in a quiet room. The right arm rested on the arm of the
25.1 1.2 yr; 19–30 yr) or an Electric Shock group (5 women,          chair with the elbow at a right angle and the wrist midway
5 men; 24.2 2.7 yr; 18–45 yr). Each subject in the Mental            between maximum supination and maximum pronation.
Math and Electric Shock groups participated in one testing              The apparatus used for the pinch task was 4 cm wide and
session that comprised a 10-min baseline and a 10-min stres-         weighed 1.2 N. It comprised a force transducer (ATI Mini-40)
sor condition. The subjects in the control group performed           mounted on a wooden platform. The sensitivity of the force
two 10-min baseline conditions. The Institutional Review             transducer was 0.06 N/V. The apparatus was held between
Board at the University of Colorado approved all experimen-          the distal pads of the thumb and index finger of the right
tal procedures, and all subjects gave their informed consent         hand. Contact surfaces of the apparatus had a rough texture
before participating in the study.                                   (sandpaper) to ensure a secure grip.
                                                                        Data from the force transducer were displayed on an
Cognitive Assessment of Anxiety                                      oscilloscope that was positioned 1 m in front of the subject at
                                                                     eye level. The target force of 4 N was presented as a line on
   To determine the general level of trait anxiety, each sub-        the oscilloscope. The subject was instructed to match the
ject completed the trait portion of the State-Trait Anxiety          exerted force with the target force as accurately as possible.
Index (STAI) of the Spielberger Index, which consists of 20          Subjects were given up to 1 min of practice with this task.
questions that are answered by using a four-point Likert-            Force was sampled at 200 Hz with the Biopac data-acquisi-
type scale. The level of state anxiety was assessed at the           tion system and stored on computer disks for later analysis.
beginning and end of the experiment with the STAI-State
(STAI-S) test and at multiple intervals throughout the ex-
periment with the visual analog mood scale (VAS). The VAS            Experimental Protocol
is a 100-mm line anchored at either end with descriptive

                                                                                                                                       Downloaded from on May 7, 2011
                                                                        The session lasted 25 min and was divided into two
polar phrases, such as “Not at all anxious” on the left side of      phases: a nonarousal baseline phase and an arousal-induc-
the 100-mm line and “Very anxious” on the right. The sub-            tion phase. For the baseline phase, the subject sat quietly for
jects were asked to place a vertical line bisecting the 100-mm       10 min and was instructed to relax. At the end of 10 min, the
line to indicate the perceived level of anxiety at the moment.       pinch task was performed for 10 s. The VAS was assessed
                                                                     three times during the baseline session: at the beginning,
Physiological Assessment of Arousal                                  middle, and end. Subsequently, subjects were either exposed
   Several variables were assessed during the experiment to          to one of the stressors (Mental Math group or Electric Shock
quantify the level of physiological arousal. Moment-to-mo-           group) or performed a second baseline period (control group).
ment changes in heart rate, blood pressure, and electroder-             Mental math stressor. Subjects in the Math group per-
mal activity were recorded by use of electrodes placed on the        formed serial subtraction of a four-digit number. In the first
left hand. Heart rate and blood pressure were measured with          5 min of the stressor, the subjects counted backward by 13,
an inflatable cuff that was placed over the proximal portion of       starting from 1,022. Subjects were instructed to count as fast
the middle finger and connected to a Finapres device (Oh-             and as accurately as possible. When a subject made a mis-
meda, Louisville, CO). Electrodermal activity was measured           take, the investigator would say, “Stop. Begin again. 1,022,”
with electrodes (Biopac, Santa Barbara, CA) that were placed         and the subject would repeat the task. To increase the diffi-
on the distal pads of the fourth and fifth fingers. Electroder-        culty of the task, subjects were asked to keep pace with a
mal activity measures the bioelectrical characteristics of the       metronome that produced an auditory signal once every 3 s.
skin by applying a direct current to the skin and recording          After 5 min, the subject was told to stop counting and to
the output as skin conductance (4). The flow of current along         perform the pinch task and complete a VAS assessment.
the skin is influenced by the activity of the eccrine sweat           Subsequently, the math task was continued for another 5
glands, which are located primarily in plantar and palmar            min. The subject was told, “This task is obviously too difficult
surfaces of the feet and hands. Because these sweat glands           for you. Instead, please count backward by 7, again starting
are innervated by the sympathetic branch of the autonomic            from 1,022. Begin.” Immediately after the second bout of the
nervous system, the change in skin conductance is often used         math stressor, the 10-s pinch task was repeated, and the VAS
as an index of sympathetic activity. A Biopac data-acquisi-          was assessed. The subject was then instructed to sit quietly
tion system was used to collect the electrodermal activity,          for a 2-min recovery period, after which the VAS, state
heart rate, and blood pressure data. The data were sampled           anxiety (STAI-S), and salivary cortisol assessments were
at 200 Hz and stored on computer disks for later analysis.           performed.
   In addition to these moment-to-moment measures, sali-                Electric shock stressor. The Shock group followed a similar
vary cortisol samples, which are a measure of adrenal output         protocol to the Math group, except that a variable-intensity
of free cortisol (24, 36), were collected at the beginning and       electric shock was used as the stressor. After the 10-min
end of the session, similar to procedures used by Kirschbaum         baseline (nonarousal) condition, two 1       1-cm carbon elec-
et al. (24). Each subject chewed a cotton swab for 20 s, which       trodes were attached to the dorsal surface (back) of the left
was then placed in a salivette for later analysis. Cortisol          hand. The electrode leads were attached to an electric stim-
levels were assessed using an enzymatic immunosolvent as-            ulation device (Grass Instruments, Quincy, MA) that was
say (Diagnostic Systems Laboratories, Webster, TX) to deter-         used to deliver a strong shock to the hand. The shock was
mine levels of free cortisol. Previous studies have established      delivered as a twin-pulse rectangular waveform with a 0.2-s
the reliability of this technique (13, 37).                          duration and intensity ranging from 60 to 120 V. Two indi-
                                                                     cator lights were used to cue the subject regarding the pos-
Motor Performance Task                                               sibility of an electric shock. When a red light was on, it was
                                                                     not possible for the subject to receive the electric shock.
  The effect of heightened arousal on motor performance was          Alternatively, when a yellow light was on it was possible to
examined with a submaximal isometric pinch task, which               receive an electric shock. These no-threat and threat periods

                                        J Appl Physiol • VOL   91 • AUGUST 2001 •
                                            AROUSAL AND MOTOR PERFORMANCE                                                    823

alternated in 20-s durations. During the threat condition            parisons. Significant F statistics were followed by post hoc
(yellow light), the subject was told that it was possible to         tests to determine pairwise differences, using 95% confidence
receive up to eight electric shocks. The pattern of electric         interval methods. Data are reported as means SE in Figs.
shocks was variable across the alternating 20-s cycles; how-         1–6 and means SD in Tables 1 and 2.
ever, this pattern was consistent across subjects. To further
heighten the arousal response, the subject was told that the         RESULTS
magnitude of the electric shock would vary across repeti-
tions. Similar to the math stressor, there were two 5-min              The effects of a stressor condition on physiological
bouts. Immediately after each bout, the 10-s pinch task was          and cognitive measures of arousal and pinch task per-
performed, along with the VAS. After the 2-min recovery              formance were determined for two types of stressors
period, a salivary cortisol sample was obtained, and the             and were compared with baseline conditions and with
posttest state anxiety was assessed.
                                                                     a control group. The Math stressor consisted of paced,
   Control condition. Similar to the two experimental groups,
the control group participated in the 10-min baseline period.        serial subtraction from a four-digit number, whereas
However, this group experienced no stressors during the              the Shock stressor involved random-intensity, noxious-
second 10-min period, and, as with the stressor groups, pinch        level stimulation to the left hand. The physiological
tasks were performed at the beginning, middle, and end of            and motor performance data are presented as 10-s
each 5-min interval. At the end of the 20-min session, state         epochs at three time points during the 25-min experi-
anxiety (STAI-S), VAS, and salivary cortisol were assessed.          mental session: near the end of the 8-min baseline and
                                                                     at the midpoint and the end of an 8-min stressor.
Data Analysis                                                        Figure 1 displays representative data across each ep-
   The dependent variables for the cognitive assessment of           och for one subject in the Shock group.
anxiety were trait anxiety, before and after state anxiety,

                                                                                                                                    Downloaded from on May 7, 2011
and multiple VAS scores. The trait and state variables were          Cognitive Component of Arousal
quantified by using standardized scoring techniques (43).
The VAS data were quantified by measuring the distance, in               Trait anxiety scores did not differ significantly for
millimeters, from the left end of the 100-mm line to the point       the three groups (Fig. 2A). The scores ranged from 21
on the line where the subject marked the perceived level of          to 62, which are considered low-to-moderate levels of
anxiety. The VAS was administered eight times during the             trait anxiety. All subjects also completed a before and
experimental session: two before baseline (time points 1 and         after state anxiety survey. The scores for the control,
2); at the middle and end of the baseline (time points 3 and 4);     Shock, and Math groups at the beginning of the proto-
after giving instructions for the stressor condition, but before
                                                                     col were 27.0, 28.6, and 30.8, respectively. Subjects in
stressor initiation (time point 5); at the middle and end of the
stressor condition (time points 6 and 7); and at the end of          the control group had end scores that were lower than
recovery (time point 8).                                             the initial scores ( 5.6%), whereas the after values for
   The dependent variables for the physiological variables           the Shock and Math stressor groups were elevated
were heart rate, blood pressure, and electrodermal activity,         significantly to 27.2% and 30.0%, respectively, above
which were assessed during three 10-s epochs. These 10-s             the initial values (Fig. 2B).
intervals were after the baseline period (time point 4) and at          The cognitive component of arousal was also as-
the middle (time point 6) and the end (time point 7) of the          sessed with the VAS. Each subject completed the VAS
stressor. The three 10-s epochs were also analyzed for the           at eight time points during the experiment: four times
control subjects, except that time points 6 and 7 occurred           (time points 1 to 4) during the baseline period, three
during a second baseline period. In addition, an assessment
                                                                     times (time points 5 to 7) during the stressor, and one
of salivary cortisol was performed at the beginning and end of
the protocol in all subjects.                                        time (time point 8) at the completion of the stressor.
   The dependent variable for the pinch task was the coeffi-          There were no between-group (P           0.22) or within-
cient of variation (SD/mean force       100) of the force during     group (P 0.55) differences during the baseline period,
the 10-s intervals at time points 4, 6, and 7. Increases in the      nor were the tests significant for simple main effects on
coefficient of variation indicate a reduction in steadiness.          sex (P     0.41) or group (P      0.06). In contrast, the
                                                                     main effect of time was significant (P 0.001), as was
Statistical Analysis                                                 the time-by-group interaction when comparing the
   A two-factor ANOVA (SPSS for Windows, SPSS) was used              stressor conditions (P 0.007). The VAS scores during
to compare the dependent variables of trait and state anxiety        the stressor (time points 5, 6, and 7) for the Shock
for the three groups. A three-factor ANOVA with one re-              (31.1    5.8, 35.6    7.0, and 39.7      7.4 mm, respec-
peated-measure design (two factors between and one within)           tively) and Math groups (33.2       5.8, 45.4     7.0, and
was used to compare the dependent variables for the heart            36.1 7.4 mm, respectively) were significantly greater
rate, blood pressure, electrodermal activity, and salivary           than those for the control group (7.1      6.2, 8.6    7.5,
cortisol between the sexes and the three groups, across time         and 6.7 7.9 mm, respectively). The VAS score at time
points, and the interactions. A three-factor ANOVA with              point 5 represents anticipatory anxiety as this was
repeated measures was applied to the cognitive assessment
data of the VAS scores between the sexes and the three
                                                                     elevated after the subjects were informed about the
groups, across time points, and the interactions. A similar          stressor conditions but before the actual presentation
three-factor ANOVA with repeated measures was applied to             of the stressor. One minute after completion of the
the coefficient of variation for force between the sexes and the      stressor (time point 8), subjects in the Shock group
three groups, across time points, and the interactions. An           (25.3 5.0 mm) had significantly elevated VAS scores
alpha level of 0.05 was chosen for all initial statistical com-      compared with the control subjects (5.1 5.3 mm) (P
                                        J Appl Physiol • VOL   91 • AUGUST 2001 •
824                                             AROUSAL AND MOTOR PERFORMANCE

                                                                                                                                    Downloaded from on May 7, 2011
Fig. 1. Representative records of heart rate (HR), systolic blood pressure (SBP), diastolic blood pressure (DBP),
electrodermal activity (EDA), and pinch-grip force for 1 subject in the Shock group. Target force for the pinch grip
was 4 N. Records represent 10-s epochs during the baseline period (10 min) and at the middle and end of the
stressor presentation. Total duration of the experiment was 25 min. bpm, Beats/min.

0.001) (Fig. 3A). However, the time-by-sex interaction                   simple main-effects test for group was significant (P
was not significant (P 0.54). The women and men in                        0.04), with electrodermal activity for the Shock group
the two stressor groups combined (Fig. 3B) and for the                   (75.3   15.9%) significantly different from the control
Shock group (Fig. 3C) and Math groups alone had                          group (12.7 16.9%), whereas the values for men and
similar VAS scores. Taken together, these data indi-                     women were similar (P 0.28).
cate that subjects in the two stressor conditions expe-                    The results were similar for systolic blood pressure,
rienced greater levels of cognitive arousal.                             with a significant main effect for time (P     0.01) and
                                                                         sex (P    0.02), whereas the main effect for group was
Physiological Arousal                                                    not significant (P    0.08) (Table 1). In addition, there
  Heart rate, systolic blood pressure, diastolic blood                   was a significant time-by-group interaction, with the
pressure, and electrodermal activity were measured at                    end-stress value for the Shock group significantly dif-
three time points during the study: baseline (time point                 ferent compared with both the baseline value for the
4), midstress (time point 6), and end-stress (time point                 Shock group and the end-stress time point for the
7) (Fig. 4). For heart rate, the simple main-effect test                 control group (Table 2). When expressed as percent
for groups was not significant (P 0.32), although the                     change (Fig. 4C), the main effects test for group was
main-effects test for time (P       0.02) and sex (P                     significant (P     0.04), with the Shock group (11.8
0.001) were significant. Heart rate at the end-stress                     3.1%) exhibiting elevated values compared with the
time point (70.9      1.7 beats/min) was significantly                    control group ( 0.3      3.3%), whereas values for men
elevated over baseline (67.4 2.0 beats/min) and mid-                     and women were similar (P         0.06). There were no
stress (68.3      1.8 beats/min) time points, whereas                    significant changes in diastolic blood pressure (Fig.
average values for women (76.3 2.4 beats/min) were                       4D).
elevated compared with average values for the men                          In addition, salivary cortisol was assessed immedi-
(61.4 2.5 beats/min) (Table 1).                                          ately before and after the session. No group, sex, or
  Simple main effects for sex (P 0.27) and group (P                      time differences were noted (Tables 1 and 2).
0.38) were not significant for electrodermal activity. In                 Pinch Task Performance
contrast, the simple main effect for time (P      0.001)
was significant, with the midstress value (113.7 42.0                       The submaximal pinch task was performed at mul-
S) being significantly greater than baseline (83.8                        tiple times during the baseline and stressor condi-
37.6 S) (Table 1); however, the end-stress value                         tions. Steadiness during the task was quantified as
(113.0     47.7 S) was not significantly different from                   the coefficient of variation for force. Accordingly,
baseline (P 0.06). Similarly, when the data are pre-                     increases in the coefficient of variation indicate a
sented as percent change from baseline (Fig. 4B), the                    reduction in steadiness. Fluctuations in the force
                                            J Appl Physiol • VOL   91 • AUGUST 2001 •
                                                   AROUSAL AND MOTOR PERFORMANCE                                                          825

                                                                                                                                                  Downloaded from on May 7, 2011
Fig. 2. Trait and state anxiety of the subjects in the 3 groups. A: trait
anxiety scores based on the Spielberger State-Trait Anxiety Index. B:
percent change in state anxiety score from beginning to end of the
experiment. Values are means         SE. *Subjects in the 2 stressor
groups experienced a significant increase in state anxiety compared
with the control group (P 0.03). % , percent change.

during the pinch task (Fig. 1) varied across groups,
sex, and time. There were no significant differences
in coefficient of variation during the baseline period
across groups: 1.65     0.60, 1.65   0.51, and 2.15
0.52% for the control, Shock, and Math groups, re-
spectively. Although the main effects for time (P
0.01) and group (P 0.02) were significant, the main
effect for sex was not significant (P        0.10). The
coefficients of variation for mid- and end-stress time                       Fig. 3. Scores on the visual analog scale (VAS) for the 3 groups of
points (2.3    0.2 and 2.8    0.3%, respectively) were                      subjects at 8 time points during the 25-min experiment. Data are
                                                                            means      SE. Time points 1 to 4 were taken during the 8-min
significantly different from baseline values (1.8                            baseline period. Time points 6 and 7 correspond to the middle and
0.1%), whereas the coefficient of variation for the                          end of the stressor presentation, respectively. A: VAS scores for
Shock group (3.01 0.33%) was significantly greater                           the Shock and Math groups were significantly greater than those
than that for the control group (1.58     0.35%) (Fig.                      for the control group at time points 5, 6, and 7 (*P 0.01), and the
                                                                            score for the Shock group remained elevated at time point 8 (*P
5A). Similarly, there were significant differences be-                       0.01). B: men and women in the Math and Shock groups (n          10
tween groups at the mid- and end-stress time points.                        in each group). C: men and women in the Shock group (n         5 in
The coefficients of variation at these times for the                         each group).
Shock group (3.25     0.38 and 4.33     0.57%, respec-
tively) were elevated compared with baseline (1.64
                                              J Appl Physiol • VOL   91 • AUGUST 2001 •
826                                              AROUSAL AND MOTOR PERFORMANCE

Fig. 4. Percent change from baseline
at the middle and end of stressor pre-
sentation for the 3 groups. Data are
means     SE. A: heart rate. B: electro-
dermal activity. C: systolic blood pres-
sure. D: diastolic blood pressure.

                                                                                                                                           Downloaded from on May 7, 2011
0.17%), along with a significant increase over each                         ence between the Math and the control or Shock
corresponding time point compared with the control                         groups. These data indicate that subjects exposed to
group (1.65      0.18, 1.55      0.41, 1.52     0.61%,                     the electric shock stressor experienced a reduction in
respectively) (P    0.001) (Fig. 5B). The sex-by-time                      steadiness during the pinch task.
interaction was not significant (P      0.05) (Fig. 5C),                       There were several statistical interactions involving
whereas the sex-by-group interaction was significant                        differences between the men and women, including a
(P     0.01) (Fig. 5D). In addition to a significant                        significant time-by-sex interaction (P     0.04) (Fig. 5C).
elevation from baseline (1.64 0.17%), the mid- and                         Post hoc analysis indicated that the coefficient of varia-
end-stress values (3.25      0.38 and 4.33      0.57%,                     tion values for the women at the end-stress interval
respectively) for the Shock group were significantly                        (3.45    0.47%) were significantly greater than baseline
elevated over the corresponding time points for the                        (1.81 0.14%) but were not different from the men at the
control group (1.55 0.41 and 1.52 0.61%, respec-                           end-stress interval (2.16 0.49%). In contrast, the coef-
tively) (Fig. 5D). The change in coefficient of varia-                      ficient of variation did not vary across time for the men.
tion for the Math group did not differ statistically                       The other significant sex effect involved the sex-by-group
across the three time points, nor was there a differ-                      interaction (P 0.04) (Fig. 5D). The coefficient of varia-

Table 1. Comparison of physiological data for women and men and across all time points
                          Heart Rate,             Systolic Blood                 Diastolic Blood        Electrodermal       Salivary
                           beats/min             Pressure, mmHg                 Pressure, mmHg           Activity, S      Cortisol, g/dl

  Men                    61.4   10.6              136.5     18.5                  77.2    9.9            94.9   42.6       1.27   0.78
  Women                  76.3   9.2*              120.4     18.7*                 70.5    9.2           111.9   41.7       1.95   1.60
  Baseline               67.4   12.9              123.1     20.2                  72.0    10.1           83.8   37.6       1.62   1.31
  Midstress              68.6   12.1              131.7     19.9†                 75.2    9.9           113.7   42.0†
  End-stress             71.3   12.0              130.4     20.8                  74.7    10.2          113.0   47.7       1.55   1.37
  Control                65.5   8.9               121.8     15.6                  72.8    10.6           96.0   35.7        1.6   1.3
  Shock                  72.3   14.0              136.7     18.6                  76.2    11.8          118.6   45.2       1.42   0.5
  Math                   68.9   12.1              127.0     21.8                  72.7    8.3            95.7   47.7       1.81   1.9
  Data are means      SD. * P     0.05, vs. men; † P      0.05, vs. baseline.

                                             J Appl Physiol • VOL    91 • AUGUST 2001 •
                                            AROUSAL AND MOTOR PERFORMANCE                                                             827

Table 2. Comparison of physiological data by group and time
                     Heart Rate,             Systolic Blood               Diastolic Blood          Electrodermal               Salivary
                      beats/min             Pressure, mmHg               Pressure, mmHg             Activity, S              Cortisol, g/dl

  Baseline           65.7   9.6              121.6   17.0                  72.8     7.3             87.7   31.2               1.47   1.05
  Midstress          66.2   7.2              123.5   16.5                  73.9     11.3           106.3   38.4
  End-Stress         66.4   8.9              117.2   15.7                  71.9     10.6            92.0   35.7               1.86   1.46
  Baseline           72.3   14.9             128.5   23.3                  73.8     11.9            85.2   31.9               1.61   1.24
  Midstress          70.0   13.4             140.6   19.9                  77.3     10.3           132.3   37.2
  End-Stress         74.4   14.0             141.1   18.6*                 77.4     11.8           138.2   45.2*              1.14   0.46
  Baseline           64.8   13.3             119.1   20.4                  69.5     10.9            78.8   49.7               1.79   1.70
  Midstress          69.3   15.0             130.2   20.9                  74.1     8.8            101.7   46.8
  End-Stress         72.5   12.1             131.8   21.8                  74.4     8.3            106.7   52.0               1.70   1.91
 Data are means   SD. * P    0.05 vs. baseline.

tion values for women (4.2 0.47%) in the Shock group                 0.47%) in the Math group (Fig. 5D). In contrast, the
were significantly greater than those for the women                   three-way interaction between sex, group, and time was
(1.67    0.47%) and men (1.49     0.52%) in the control              not significant (P 0.06).
group, the women in the Math group (2.09 0.47%), and                   When the coefficient of variation was expressed as a

                                                                                                                                              Downloaded from on May 7, 2011
the men (2.03 0.47%) in the Shock group (P 0.04) but                 percent change from baseline for the two stressor time
did not differ significantly from values in men (2.47                 points, the main effects tests were significant for

                                                                                                  Fig. 5. Coefficient of variation for force
                                                                                                  during the pinch-grip task. A: main
                                                                                                  effect for group, showing the data
                                                                                                  summed across the baseline, mid-
                                                                                                  stress, and end-stress time points.
                                                                                                  *Significant difference between Shock
                                                                                                  and control groups (P          0.02). B:
                                                                                                  group time interaction, showing the
                                                                                                  data at the baseline (time point 4), mid-
                                                                                                  stress (time point 6), and end-stress
                                                                                                  (time point 7) intervals. *End-stress
                                                                                                  and baseline intervals for the Shock
                                                                                                  group are significantly different (P
                                                                                                  0.01); #significant difference between
                                                                                                  the end-stress interval and all inter-
                                                                                                  vals for the control group (P 0.02). C:
                                                                                                  sex     time interaction, showing the
                                                                                                  data at baseline, midstress, and end-
                                                                                                  stress intervals. D: sex    group inter-
                                                                                                  action. *Significant difference for
                                                                                                  women in the Shock group compared
                                                                                                  with women in the control and Math
                                                                                                  groups (P      0.01); #significant differ-
                                                                                                  ence between women in the Shock
                                                                                                  group and men in the Shock and con-
                                                                                                  trol groups (P 0.02).

                                        J Appl Physiol • VOL   91 • AUGUST 2001 •
828                                             AROUSAL AND MOTOR PERFORMANCE

                                                                         group, sex, and time. The Shock group (121.3 18.1%)
                                                                         differed significantly from the control ( 2.5    19.2%)
                                                                         and Math (8.8     18.1%) groups (P    0.001) (Fig. 6A).
                                                                         Similarly, women increased by 67.3 14.8% compared
                                                                         with 17.8 15.4% for the men (P 0.03) (Fig. 6B), and
                                                                         the end-stress time point (56.9 14.6%) was elevated
                                                                         from the midstress time point (28.2 8.7%) (P 0.02)
                                                                         (Fig. 6C).


                                                                            The purpose of this study was to determine the effect
                                                                         of arousal in men and women on the moment-to-mo-
                                                                         ment performance of a simple motor task. Subjects
                                                                         performed a series of submaximal pinch tasks during a
                                                                         baseline period and during a stressor condition. We
                                                                         expected to find that the stressor conditions would
                                                                         elevate arousal and result in a reduction in the accu-
                                                                         racy of the pinch task. The main finding was that
                                                                         subjects in the stressor groups, especially the Shock
                                                                         group, displayed significantly reduced steadiness at

                                                                                                                                     Downloaded from on May 7, 2011
                                                                         the midstress and end-stress time points. Further-
                                                                         more, women in the Shock group exhibited more of an
                                                                         impairment than men. This selective effect on perfor-
                                                                         mance was in contrast to the common effects of the
                                                                         interventions on the arousal of the men and women in
                                                                         the two stressor groups.
                                                                            Previous studies have used the mental math and
                                                                         shock stressors to heighten arousal. For example, Kirsch-
                                                                         baum et al. (24) combined mental arithmetic and sim-
                                                                         ulated public speaking to induce moderate psychologi-
                                                                         cal stress, as indicated by elevated heart rate and
                                                                         salivary cortisol levels. Similarly, Al’Absi and col-
                                                                         leagues (1) demonstrated elevated systolic and dia-
                                                                         stolic blood pressure, heart rate, and salivary cortisol
                                                                         with mental arithmetic and simulated public speaking.
                                                                         However, all physiological measures were attenuated
                                                                         with repeated bouts of the same stressor. In addition to
                                                                         measuring neuroendocrine and cardiovascular re-
                                                                         sponses, Breznitz et al. (5) asked subjects to rate their
                                                                         level of tension at multiple intervals during shock
                                                                         stressors. Tension, as rated with a seven-point scale,
                                                                         was significantly elevated during the shock stressor
                                                                         compared with baseline, as were heart rate and plasma
                                                                         epinephrine levels.
                                                                            We used two assessments of the cognitive component
                                                                         of arousal, the Spielberger state anxiety index and
                                                                         VAS. The anxiety scores at the end of the test were
                                                                         significantly elevated compared with those before the
                                                                         test for the two stressor groups, whereas scores for the
                                                                         control group actually decreased. Similarly, the VAS
Fig. 6. Percent change in the coefficient of variation for force during
the pinch-grip task, with main effects for group, sex, and time. A:      scores for the Math and Shock groups were elevated
normalized fluctuations for the 3 groups of subjects, collapsed across    significantly compared with the control group scores
sex and the mid- and end-stress intervals, with values for the Shock     during the three stressor intervals. Although the state
group significantly greater than those for the Math and control           anxiety index has frequently been used to assess tem-
groups (*P      0.001). B: normalized fluctuations for sex, collapsed
across the 3 groups and the mid- and end-stress intervals, with
                                                                         poral changes in state anxiety (14, 40, 42), the time
values for the women significantly greater than the values for the        required to complete the 20-item questionnaire makes
men (*P      0.03). C: normalized fluctuations at the mid- and end-       it an impractical tool for the assessment of moment-to-
stress intervals, collapsed across sex and groups, with values at the    moment changes in anxiety. In contrast, the VAS can
end-stress interval significantly greater than those at midstress         be used to assess emotional states rapidly. In a reli-
(*P 0.02). Values are means SE.
                                                                         ability study of the Spielberger state anxiety test and
                                            J Appl Physiol • VOL   91 • AUGUST 2001 •
                                        AROUSAL AND MOTOR PERFORMANCE                                                 829

VAS, Cella and Perry (7) reported that the VAS mea-               of either the Shock or control groups. The coefficient of
sure of anxiety was significantly correlated (r       0.53)        variation for force has been used in other studies as an
with the state anxiety scores and concluded that the              index of fine motor control (11, 26). In the present
VAS is capable of measuring emotional states in a                 study, the coefficient of variation ranged from 1.4% for
quick, reliable, and relatively sensitive manner. We              men in the control group during baseline to 6.1% for
found that state anxiety immediately after the stressor           women in the Shock group at the end-stress time point,
increased by       30% for the two stressor groups,               which is similar to the range reported previously (11).
whereas the VAS scores at the three time points during               Electric shock has been used to examine the startle
the stressor conditions increased by 200% from base-              response. Grillon and colleagues (14–16) used a light-
line values.                                                      ing system to provide threat and nonthreat conditions
   In addition to cognitive assessment, physiological             while delivering noxious levels of shock. Electric shock
markers were used to measure the arousal response.                potentiated the acoustic startle response, elevated per-
Generally, the two stressor conditions produced simi-             ceived tension, and enhanced reflex responses. Simi-
lar changes in heart rate, systolic and diastolic blood           larly, inescapable tail shock has been used to induce
pressure, electrodermal activity, and salivary cortisol.          acute and chronic stress in rats, resulting in long-term
However, diastolic blood pressure and cortisol mea-               neural, behavioral, and immune changes (9, 32). Sev-
sures were not different compared with the control                eral neural mechanisms may contribute to the associ-
group, as found previously (1, 12). Although salivary             ation between electric shock and altered motor output.
cortisol is frequently used as a marker of physiological          Chua et al. (8), for example, identified a number of
arousal, the delayed time course may explain the ab-              paralimbic structures that were active during and im-
sence of group or time differences in this study. For             mediately after unpredictable electric shock, perhaps

                                                                                                                             Downloaded from on May 7, 2011
example, Kirschbaum et al. (24) noted peak cortisol               mediating changes in arousal associated with anticipa-
levels to occur at 40 min after a laboratory stressor,            tory states. This likely includes the amygdala, which is
compared with the 25-min duration between our initial             involved in the initiation and maintenance of inte-
and final measurements. The brevity of our testing                 grated stress responses that include projections to ar-
session likely explains the minimal changes in salivary           eas affecting stress hormones via anterior pituitary/
cortisol values we observed. Other authors using sim-             adrenal cortex, sympathetic activity, and descending
ilar testing procedures have demonstrated good reli-              motor activity to the ventromedial horn of the spinal
ability for within- and between-day assessments (13,              cord. By influencing the prefrontal-striatal system at
37).                                                              both the cortical and striatal levels, the amygdala is
   Increased neuroendocrine activity, especially en-              profoundly involved in motor, cognitive, and complex
hanced activation of the sympathetic nervous system,              behavioral functions (17).
can result in changes in heart rate, blood pressure, and             Although men and women had similar arousal re-
electrodermal activity (6, 24). Stressor exposure also            sponses to the shock stressor, the reduction in steadi-
activates the hypothalamus-pituitary-adrenal axis, re-            ness during the pinch task was greater for the women.
sulting in elevated cortisol. In laboratory settings,             The steadiness exhibited by women, as indicated by the
changes in these variables due to a stressor are usually          coefficient of variation for force, was reduced signifi-
small, with substantial variability between subjects.             cantly at the end-stress time point compared with
Consequently, the moment-to-moment assessment of                  baseline, whereas there was no change for men. In
multiple variables has been recommended for a more                addition, the coefficient of variation for the women in
accurate interpretation of the arousal response (31).             the Shock group was significantly different from
Al’Absi et al. (1) assessed psychological and physiolog-          women and men in the control group and for the men in
ical variables, including heart rate, blood pressure, and         the Shock group. Although no sex effect was found for
cortisol, at multiple intervals during both mental math           the Math stressor, these findings indicate that the
and public speaking stressors. Correlations between               accuracy of the pinch task was reduced in women who
the endocrine, cardiovascular, and psychological vari-            experienced the electric shock stressor compared with
ables were significant for the speaking stressor, but not          men.
the math stressor, although all values were signifi-                  There may be several explanations for the dissocia-
cantly elevated over baseline. The 19% increases                  tion between arousal and performance for the women.
from baseline for heart rate and blood pressure found             First, the measures of arousal used in this study may
by Al’Absi et al. (1) for the math stressor are similar to        not have been sensitive enough to identify sex differ-
the 14–18% increases that we found for the math and               ences. Levels of cognitive, cardiovascular, and electro-
electric shock stressors.                                         dermal activity were elevated from baseline for the
   Of the two stressors used to elevate arousal, electric         stressors using noninvasive techniques, whereas corti-
shock had the greater impact on the steadiness of the             sol levels were unchanged from baseline. However,
pinch task. The coefficients of variation for force at the         more invasive techniques may be needed to detect
midstress and end-stress time points were significantly            changes in other components of arousal that differ for
elevated for the Shock group compared with baseline               men and women. The effects of electric shock in ani-
values and for the control group. In contrast, the coef-          mals indicate that females respond with a greater level
ficient of variation for the Math group did not change             of neuroendocrine activation, including ACTH, cortico-
across the time points and was not different from that            sterone, oxytocin, norepinephrine, and serotonin re-
                                     J Appl Physiol • VOL   91 • AUGUST 2001 •
830                                             AROUSAL AND MOTOR PERFORMANCE

lease (39). Two of these neurotransmitters, serotonin                     6. Callister R, Suwarno NO, and Seals DR. Sympathetic activ-
and norepinephrine, appear to function as modulators                         ity is influenced by task difficulty and stress perception during
                                                                             mental challenge in humans. J Physiol (Lond) 454: 373–387,
of motor neuron excitability, enhancing motor neuron                         1992.
responses to excitatory input. Activation of these sub-                   7. Cella DF and Perry SW. Reliability and concurrent validity of
strates, especially serotonin and norepinephrine, could                      three visual-analogue mood scales. Psychol Rep 59: 827–833,
explain the dissociation between our arousal measures                        1986.
                                                                          8. Chua P, Krams M, Toni I, Passingham R, and Dolan R. A
and motor performance for the women. Quantifying
                                                                             functional anatomy of anticipatory anxiety. Neuroimage 9: 563–
these neuroendocrine substrates may identify the                             571, 1999.
mechanisms linking arousal and performance, espe-                         9. Deak T, Nguyen KT, Cotter CS, Fleshner M, Watkins LR,
cially those related to sex differences.                                     Maier SF, and Spencer RL. Long-term changes in mineralo-
  A second possibility may be that men and women                             corticoid and glucocorticoid receptor occupancy following expo-
                                                                             sure to an acute stressor. Brain Res 847: 211–220, 1999.
respond to stressors with similar levels of arousal, but                 10. Dickinson PS. Interactions among neural networks for behav-
tasks performed during states of increased arousal are                       ior. Curr Opin Neurobiol 5: 792–798, 1995.
characterized by altered motor output for women. For                     11. Enoka RM, Burnett RA, Graves AE, Kornatz KW, and
example, when a prolonged, submaximal fatiguing con-                         Laidlaw DH. Task- and age-dependent variations in steadiness.
traction is performed, which is a type of physical stres-                    Prog Brain Res 123: 389–395, 1999.
                                                                         12. Faulstich ME, Williamson DA, McKenzie SJ, Duchmann
sor, the longer endurance times produced after immo-                         EG, Hutchinson KM, and Blouin DC. Temporal stability of
bilization by the women can be associated with                               psychophysiological responding: a comparative analysis of men-
changes in the pattern of motor unit activation during                       tal and physical stressors. Int J Neurosci 30: 65–72, 1986.
this task (41). Although little is known about the un-                   13. Flagg D, Fleshner M, Hunter SK, and Enoka RM. Circadian
derlying sex differences in motor performance, circu-                        rhythmicity of salivary cortisol levels in men and women is
                                                                             reliable (Abstract). Med Sci Sports Exerc 32: S271, 2000.

                                                                                                                                                     Downloaded from on May 7, 2011
lating sex steroids may play a significant role (35).                     14. Grillon C, Ameli R, Foot M, and Davis M. Fear-potentiated
  In summary, subjects presented with Math and                               startle: relationship to the level of state/trait anxiety in healthy
Shock stressors experienced significant changes in                            subjects. Biol Psychiatry 33: 566–574, 1993.
cognitive and physiological arousal compared with                        15. Grillon C, Ameli R, Merikangas K, Woods SW, and Davis
                                                                             M. Measuring the time course of anticipatory anxiety using the
baseline and control subjects. Of the physiological                          fear-potentiated startle reflex. Psychophysiology 30: 340–346,
variables, heart rate, systolic blood pressure, and elec-                    1993.
trodermal activity were elevated with presentation of                    16. Grillon C and Davis M. Effects of stress and shock anticipation
the stressors, whereas diastolic blood pressure and                          on prepulse inhibition of the startle reflex. Psychophysiology 34:
salivary cortisol were unchanged. The greater levels of                      511–517, 1997.
                                                                         17. Groenewegen HJ, Wright CI, and Beijer AVJ. The nucleus
cognitive and physiological arousal were associated                          accumbens: gateway for limbic structures to reach the motor
with reductions in steadiness for the Shock subjects,                        system? In: The Emotional Motor System, edited by Holstege G,
but not for the subjects in the Math group who expe-                         Bandler R, and Saper C. Amsterdam: Elsevier, 1996, p. 499–506.
rienced heightened arousal but no change in steadi-                      18. Hoehn-Saric R, Hazlett RL, Pourmotabbed T, and McLeod
ness. Although women exhibited more of an impair-                            DR. Does muscle tension reflect arousal? Relationship between
                                                                             electromyographic and electroencephalographic recordings. Psy-
ment than men, the reduction in steadiness was                               chiatry Res 71: 49–55, 1997.
largely unrelated to the magnitude of the effect on                      19. Iversen S, Kupfermann I, and Kandel ER. Emotional states
arousal.                                                                     and feelings. In: Principles of Neural Science, edited by Kandel
                                                                             ER, Schwartz JH, and Jessell TM. Norwalk, CT: Appleton and
   We thank Dr. Robert Gotshall for the use of the Finapres device,          Lange, 2000, p. 982–997.
and Drs. Sandra K. Hunter, Douglas R. Seals, and Douglas L. Weeks        20. Jezova D, Jurankova E, Mosnarova A, Kriska M, and
for comments on a draft of the manuscript.                                   Skultetyova I. Neuroendocrine response during stress with
   This study was funded by National Institute on Aging Grant                relation to gender differences. Acta Neurobiol Exp (Warsz) 56:
AG-13929 (to R. M. Enoka) and the Foundation for Physical Therapy            779–785, 1996.
(to J. T. Noteboom).                                                     21. Katz PS. Intrinsic and extrinsic neuromodulation of motor cir-
                                                                             cuits. Curr Opin Neurobiol 5: 799–808, 1995.
                                                                         22. Katz PS. Neurons, networks, and motor behavior. Neuron 16:
                                                                             245–253, 1996.
 1. Al’absi M, Bongard S, Buchanan T, Pincomb GA, Licinio J,             23. Katz PS and Frost WN. Intrinsic neuromodulation: altering
    and Lovallo WR. Cardiovascular and neuroendocrine adjust-                neuronal circuits from within. Trends Neurosci 19: 54–61, 1996.
    ment to public speaking and mental arithmetic stressors. Psy-        24. Kirschbaum C, Pirke KM, and Hellhammer DH. The “Trier
    chophysiology 34: 266–275, 1997.                                         social stress test”: a tool for investigating psychobiological stress
 2. Arena J and Hobbs S. Reliability of psychophysiological re-              responses in a laboratory setting. Neuropsychobiology 28: 76–81,
    sponding as a function of trait anxiety. Biofeedback Self Regul          1993.
    20: 19–37, 1995.                                                     25. Kudielka BM, Hellhammer J, Hellhammer DH, Wolf OT,
 3. Bonnet M, Bradley MM, Lang PJ, and Requin J. Modulation                  Pirke KM, Varadi E, Pilz J, and Kirschbaum C. Sex differ-
    of spinal reflexes: arousal, pleasure, action. Psychophysiology 32:       ences in endocrine and psychological responses to psychosocial
    367–372, 1995.                                                           stress in healthy elderly subjects and the impact of a 2-wk
 4. Boucsein W. Electrodermal Activity. New York: Plenum, 1992.              dehydroepiandrosterone treatment. J Clin Endocrinol Metab 83:
    (Plenum Ser. Behav. Psychophysiol. Med. 442)                             1756–1761, 1998.
 5. Breznitz S, Ben-Zur H, Berzon Y, Weiss DW, Levitan G,                26. Laidlaw DH, Kornatz KW, Keen DA, Suzuki S, and Enoka
    Tarcic N, Lischinsky S, Greenberg A, Levi N, and Zinder                  RM. Strength training improves the steadiness of slow length-
    O. Experimental induction and termination of acute psycholog-            ening contractions performed by old adults. J Appl Physiol 87:
    ical stress in human volunteers: effects of immunological, neu-          1786–1795, 1999.
    roendocrine, cardiovascular, and psychological parameters.           27. Marder E. From biophysics to models of network function. Annu
    Brain Behav Immun 12: 34–52, 1998.                                       Rev Neurosci 21: 25–45, 1998.

                                            J Appl Physiol • VOL   91 • AUGUST 2001 •
                                                  AROUSAL AND MOTOR PERFORMANCE                                                             831

28. Martens R. Arousal and motor performance. Exerc Sport Sci              37. Pruessner JC, Wolf OT, Hellhammer DH, Buske-Kirsch-
    Rev 2: 155–188, 1974.                                                      baum A, von Auer K, Jobst S, Kaspers F, and Kirschbaum
29. Matthews KA, Caggiula AR, McAllister CG, Berga SL,                         C. Free cortisol levels after awakening: a reliable biological
    Owens JF, Flory JD, and Miller AL. Sympathetic reactivity                  marker for the assessment of adrenocortical activity. Life Sci 61:
    to acute stress and immune response in women. Psychosom Med                2539–2549, 1997.
    57: 564–571, 1995.                                                     38. Raglin JS. Anxiety and sport performance. Exerc Sport Sci Rev
30. Neiss R. Reconceptualizing arousal: psychobiological states in             20: 243–274, 1992.
    motor performance. Psychol Bull 103: 345–366, 1988.                    39. Rivier C. Gender, sex steroids, corticotropin-releasing factor,
31. Ney T and Gale A. A critique of laboratory studies of emotion              nitric oxide, and the HPA response to stress. Pharmacol Biochem
    with particular reference to psychophysiological aspects. In: So-          Behav 64: 739–751, 1999.
    cial Psychophysiology and Emotion: Theory and Clinical Appli-          40. Sade S, Bar-Eli M, Bresler S, and Tenenbaum G. Anxiety,
    cations, edited by Wagner HL. New York: Wiley, 1988, p. 65–83.             self-control and shooting performance. Percept Mot Skills 71:
32. Nguyen KT, Deak T, Will MJ, Hansen MK, Hunsaker BN,                        3–6, 1990.
                                                                           41. Semmler JG, Kutzscher DV, and Enoka RM. Gender differ-
    Fleshner M, Watkins LR, and Maier SF. Timecourse and
                                                                               ences in the fatigability of human skeletal muscle. J Neuro-
    corticosterone sensitivity of the brain, pituitary, and serum in-
                                                                               physiol 82: 3590–3593, 1999.
    terleukin-1 protein response to acute stress. Brain Res 859:
                                                                           42. Shostak BB and Peterson RA. Effects of anxiety sensitivity on
    193–201, 2000.                                                             emotional responses to a stress task. Behav Res Ther 28: 513–
33. Noteboom JT and Enoka RM. Electrical shock increases                       521, 1990.
    physiological arousal and impairs motor performance (Abstract).        43. Spielberger CD, Gorsuch RL, and Lushene RE. State-Trait
    Med Sci Sports Exerc 32: S282, 2000.                                       Anxiety Inventory Manual. Palo Alto, CA: Consulting Psycholo-
34. Ogilvie KM and Rivier C. Gender difference in hypothalamic-                gists Press, 1970.
    pituitary-adrenal axis response to alcohol in the rat: activational    44. Spielberger CD and Rickman RL. Assessment of state and
    role of gonadal steroids. Brain Res 766: 19–28, 1997.                      trait anxiety. In: Anxiety: Psychobiological and Clinical Perspec-
35. Patchev VK and Almeida OF. Gender specificity in the neural                 tives, edited by Sartorius N. New York: Hemisphere, 1990, p.
    regulation of the response to stress: new leads from classical             69–83.

                                                                                                                                                    Downloaded from on May 7, 2011
    paradigms. Mol Neurobiol 16: 63–77, 1998.                              45. Weinberg R and Ragan J. Motor performance under three levels
36. Peters ML, Godaert GLR, Ballieux RE, van Vliet M, Willem-                  of trait anxiety and stress. J Motor Behav 10: 169–176, 1978.
    sen JJ, Sweep FC, and Heijnen CJ. Cardiovascular and endo-             46. Yerkes RM and Dodson JD. The relation of strength of stim-
    crine responses to experimental stress: effects of mental effort and       ulus to rapidity of habit formation. J Compar Neurol Psych 18:
    controllability. Psychoneuroendocrinology 23: 1–17, 1998.                  459–482, 1908.

                                              J Appl Physiol • VOL   91 • AUGUST 2001 •

hkksew3563rd hkksew3563rd http://