Download Effects of Psychological and Physiological Challenges on Heart

Back to Psychophysiological
Home Page
Internationol Journal of Psychophysiology, 22 (1996) pp. 173-183.
TWave_pulse96.doc
Effects of psychological and physiological challenges on heart rate,
T-wave amplitude, and pulse-transit time
John J. Furedy a,*, Attila Szabo b,*, Francois Peronnet c
a
Department of Psychology, University of Toronto, Toronto, Ontario, Canada M5S IA
Department of Health Science and Sports Medicine, Hungarian University of Physical Education, Alkotcus u. 44, H-1123 Budapest,
Hungary
c
Departement d’education physique, Universite de Montreal, CP6128, succursale Centre-Ville, Montreal, Quebec. Canada H3C 3J7
b
Received 26 October 1995; revised 16 February 1996; accepted 16 February 1996
Abstract
The reactive sensitivities of T-wave amplitude (TWA), pulse-transit time (PTT), and heart rate (HR) were
examined in response to psychological, physiological, and combined challenges. In one experiment, 20
subjects performed I-min arithmetic and combined arithmetic-with-cycling tasks, with HR and TWA being
measured. The former showed significant reactive sensitivity, but TWA attenuation reached significance only
in the combined challenge situation. In another experiment, 18 males performed 1 min arithmetic tasks,
before, during, and following sustained low and moderate intensity cycling. Pulse-transit time was also gauged
in this second study. The results showed that HR increased reliably to all challenges, TWA attenuated in
response to the arithmetic task both at rest and during exercise, but displayed the paradoxical augmentation to
sustained exercise, and PTT decreased significantly to exercise, but it did not decrease reliably to the
arithmetic task in any of the conditions. These results suggest that the time course (tonic versus phasic) of the
challenge, rather than its psychological or physiological nature, may be the determinant factor in TWA
reversal.
Keywords: Arithmetic challenge: Exercise: Heart rate; Pulse-transit time; Reactive sensitivity; T-wave amplitude
1. Introduction
In general, psychophysiological studies of cardiac
reactivity to challenges are mainly concerned with the
sympathetic component of the autonomic reactivity rather
than with the parasympathetic component, because it is
the former component that is more central to the
concept of stress. Heart rate (HR) increase, which is the
most convenient cardiac per-
* Corresponding authors.
formance measure, fails as a candidate index of
myocardial sympathetic activity on grounds of 'specific'
index sensitivity (see Furedy and Heslegrave, 1984),
because it is supra-ventricular in origin, and is, therefore,
significantly affected by parasympathetic influences. On
the other hand, changes in the electrocardiographic Twave amplitude (TWA) and in the contractile index of
pulse-transit time (PTT) are both based on ventricular
functions. However, aside from relatively complex and
disputed theoretical predicaments with both measures
(Contrada, 1992; Contrada et al., 1991; Furedy, 1987;
Furedy et al., 1992; Heslegrave and
174
J.J. Furedy et al./Internationol Journal of Psychophysiology 22 (1996) 173-183
Furedy, 1980) there is also evidence for a lack of reactive
sensitivities (as compared to HR for example) of these
indices in various challenging situations.
The problem with TWA is that, depending on the
challenge, it shows bidirectional responses. For example,
TWA decreases very consistently in response to mental
challenge (Contrada et al., 1989; Hijzen et al., 1984; Scher
et al., 1984), but it shows a paradoxical augmentative
response to sustained physical exercise (Blomquist, 1969;
Hartung, 1972, Hartung and Nouri, 1979; Mayhew, 1971;
Rose et al., 1966). Further, in response to
pharmacological challenge TWA may decrease or
increase depending on the dose of the sympathomimetic
agent (Contrada et al., 1991). A different problem for PTT
is that psychological challenges, like some forms of mental
arithmetic (MA) tasks, do not elicit detectable responses
even though they yield TWA attenuation and/or HR
acceleration (Linden and Estrin, 1988; Roth et al., 1990;
Szabo, 1993).
The purpose of the two studies reported here was to
provide further information on the reactive-sensitivity
problems of TWA and PTT. Taken together, these
experiments allowed us to compare the two candidate
indices, along with HR, at rest and during light and
moderate exercise, during combined psychological and
physiological challenges, and during phasic and tonic
challenges. The comparisons were expected to shed light
on the specificities and sensitivities of TWA and PTT
under three forms of stimulus characteristics: (1) in a
physical condition, (2) in a psychological condition, and
(3) in a combined physical and psychological condition.
Further, based on the frequently observed reversal in TWA
during prolonged exercise, the comparisons were expected
to elucidate whether such a reversal is intensified or
counteracted by the addition of a psychological challenge
to the ongoing physical challenge. Based on the current
status of the literature, it was predicted that TWA will
decrease in response to psychological challenge in all
situations even when a reversal due to prolonged exercise
has taken place. With regard to PTT, it was expected that
this measure will show little change in response to psychological challenge, but that it will decrease during light
and moderate exercise. In Experiment II, calculation of
the effect sizes (Cohen, 1969) was intro-
duced to determine the significance of the changes (if
any) in TWA and PTT in comparison to HR. While
statistical significance may demonstrate change, the
effect size is a more accurate measure of the 'magnitude' of
change and may be a useful index of the sensitivity of the
three dependent variables under the various stimulus
conditions. It was conjectured that this method would help
in the comparison of the sensitivity and specificity of the
dependent measures.
2. Experiment I
The first experiment was performed to examine the
TWA responses to phasic psychological challenge alone
as well as in combination with phasic physical challenge.
The latter condition was expected to clarify whether TWA
decrease (expected to occur in response to MA) will be
enhanced or opposed by the addition of the physical
challenge. The answer was expected to provide
information about the specificity and sensitivity of TWA.
Heart rate was used as the 'criterion' reactive-sensitivity
measure.
2.1. Methods
2.1.1. Subjects
Subjects were recruited from a large university milieu.
The participants answered the Physical Activity Readiness
Questionnaire (PAR-Q, British Columbia Ministry of
Health, 1978) for safety reasons. Twenty nonsmoker
males (mean age = 26.8 years, SD = 6.0; mean height
=178.1 cm, SD = 5.8; mean weight = 74.8 kg, SD = 7.7)
volunteered for the study.
2.7.2. Materials
Heart rate and TWA were recorded by using Ag/AgCl
(Medi-Trace) electrodes coupled to a Cambridge (Model
VS-550) electrocardiogram via a Marcom (Model 500061-H) oscilloscope voltage transmitter. An automated
timer-recorder (Quinton) was used for data logging at preset intervals. Cycling was performed on a Monark
(Model 668) stationary bicycle. A digital revolution
(RPM) counter (Lafayette Instruments, Model 54419-A)
was used to control the cycling pace. Habituation to
J.J. Furedy et al./Intemational Journal of Psychophysiology 22 (1996) 173-183
cycling rhythm was aided by using a Taktell Junior
(Model 826) Metronome.
2.1.3. Mental arithmetic
The mental challenge was designed on the basis of
previous work (Szabo and Gauvin, 1992a). In short,
participants were asked to solve four MA problems, of
approximately equal difficulty level, in 1 min. Each
problem consisted of three basic operations involving
single, double, and triple digits (e.g. 189 X 2 + 10 - 7).
The problems were presented visually, required verbal
answers, and could be solved in any order of preference to
stimulate subjects' performance (Szabo and Gauvin,
1992b). Since it is known that vocalization affects HR
(Brown et al., 1988), the answers given (and hence
vocalizations) were tape-recorded for the later analysis
of any possible between-condition differences.
2.1.4. Procedure
Upon arrival to the laboratory, the protocol was
explained both verbally and in writing to the subjects.
Next, electrodes were placed on subjects' thorax in a
modified lead-l arrangement. Following trial ECG
recordings, subjects were seated in an armchair, the
arithmetic task was explained, and a trial problem was
presented to habituate subjects with the mental task. After
a 10 min rest period in the armchair, subjects were led to
the bicycle, the height of the bicycle seat was adjusted to
the subjects' height and the subjects were asked to sit on
the bicycle by placing their foot on a lower handle while
holding the upper handle with both hands. A resting period
in this body-posture was implemented to allow the
stabilization of the HR. Heart rate was considered stable
when its fluctuation did not exceed five beats per min
within 60 s. The interval for HR stabilization was less than
5 min in all instances. Immediately after a resting ECG
recording the subjects were presented with three,
consecutive, 1 min MA tasks. During the first and the third
minute the subjects performed the mental task while
simply sitting on the bicycle. However, during the second
minute the subjects were also asked (beside doing the
MA problems) to cycle at approximately 60 W for the
duration of the minute. The ECG recordings were taken
in the last 10s of five consecutive minutes (i.e.
anticipation, MA task before cycling, MA task +
175
cycling, MA task after cycling, no-task). Heart rate and
TWA were determined from the ECG chart. For the sake
of convenience the five 1 min sampling periods in
Experiment I will be referred to as: (1) anticipation (sitting
on the bicycle before treatment), (2) math-l (MA while
sitting on the bicycle before the cycling), (3) math +
cycling (MA while cycling at about 60 W), (4) math-2
(MA while sitting on the bicycle after the cycling), and (5)
no-tusk (sitting on the bicycle after the treatment).
2.2. Results
A repeated measure analysis of variance
(ANOVA) in the five consecutive conditions (i.e.
anticipation, math-l, math + cycling, math-2, and no-task)
of the HR data was significant (F (4,76) = 72.6, p<
0.001; Greenhouse-Geisser (G-G) p< 0.001, E =
0.4920; Fig. 1). Simple contrasts (Wilkinson, 1989)
showed that HR during anticipation was lower than HR
during math-1 (F (1,19) = 26.7, p < 0.001), than HR
during math + cycling (F(1,19) = 88.4, p < 0.001), and
than HR during math-2 (F (1,19)= 11.0, p< 0.004), but
not significantly different from HR during no-task. When
cycling was superimposed on the MA task, HR was
significantly higher than during math-l (F (1,19) =
88.2, p< 0.001) and during math-2 (F (1,19)= 114.3, p
< 0.001). However, math-2 HR was lower than math-l HR
(F (1,19) = 10.4, p < 0.004).
A similar repeated measure ANOVA of TWA was
also significant (F (4,76) = 33.0, p < 0.001; G-G p<
0.001, E = 0.7267; Fig. 2). Simple contrasts revealed that
TWA was significantly lower, in comparison to
anticipation, only during math + cycling (F (1,19)= 14.5,
p< 0.001). From anticipation to math-l there was a trend
toward lower TWA, but it did not reach the acceptable
statistical significance (F (1,19) = 3.7, p < 0.07). The
TWA was lower during math + cycling than both
during math-l (F (1,19) = 37.0, p < 0.001) and math-2
(F (1,19) = 52.9, p< 0.001). The TWA during math-2
was greater than during math-l (F (1,19) = 17.8, p<
0.001). Moreover, during no-task TWA was greater than
during anticipation (F (1,19) = 17.8, p < 0.001). Finally,
during math-2 TWA was not different from the TWA
during no-task.
176
J.J. Furedy et al./Internationol Journal of Psychophysiology 22 (1996) 173-183
Fig. 1. Heart rate in five (consecutive) experimental conditions:
(1) anticipation (simply sitting on the bicycle), (2) math-l (sitting
on the bicycle and performing MA), (3) math + cycling, (4)
math-2 (no cycling, doing MA while sitting on the bicycle), acd
(5) no-task (simply sitting on the bicycle as during
anticipation).
2.3. Discussion
The HR results (Fig. 1) confirmed the reactive
sensitivity of this measure, inasmuch as it reliably
increased to the psychological challenge and yielded
a significantly greater increase to the combination of the
psychological and physiological challenges. The T-wave
attenuation results paralleled the HR results (Fig. 2), and
showed no evidence for the paradoxical augmentative
effect to the physiological challenge (exercise) used in this
study. However, as opposed to HR, TWA increased
following the combined challenge situation as based on the
comparison of the anticipation phase with the no-task phase
and math-l with math-2 values. One explanation for this
phenomenon may be associated with a post-exercise
relaxation (i.e. sympathetic withdrawal or rebound)
induced by brief exercise. Such an explanation could also
account for the lack of differences between TWAs
during math-2 and no-task. Another possible explanation
may be that in the anticipation phase TWA was already
decreased (as compared to the no-task phase at the end)
due to some expectations, related to cycling or mathanxiety, that could be psychologically challenging. If the
latter is the case, it may also explain why the attenuation
from the anticipation to math-l was only marginally
significant.
3. Experiment II
In addition to HR and TWA, PTT was also recorded
as a dependent variable in the second experiment. The
psychological 1 min MA challenge was retained, but it
was combined with a longer-duration sustained
physiological exercise challenge, which was also varied
between light to moderate (40% maximal heart rate
reserve-MHRR) and moderate to high (60% MHRR)
intensity levels. Finally, we also added a number of
performance and self-report measures, to see whether
there is a parallel between these measures and the more
objective psychophysiological measures.
3.I. Methods
3.1.1. Subjects
Participants for Experiment II were graduate students at
University of Montreal. All subjects were screened for
healthy drug- and smoke-free lifestyle. They reported
participation in an average of three vigorous bouts of
physical activities per week in the
Fig. 2. T-wave amplitude in five (consecutive) conditions: (1)
anticipation (simply sitting on the bicycle), (2) math-l (sitting on
the bicycle and performing MA), (3) math+cycling, (4) math-2
(no cycling, doing MA while sitting on the bicycle), and (5)
no-task (simply sitting on the bicycle as during anticipation).
J.J. Furedy et al./Intemational Journal of Psychophysiology 22 (1996) 173-183
177
past 4 months on the Exercise Behavior Questionnaire
(Godin et al., 1986) and no contraindications to exercise
on the PAR-Q (British Columbia Ministry of Health,
1978). The final sample consisted of 18 males (age =
29.0 years, SD = 6.7; height =177.6 cm, SD = 5.7;
weight = 74.7 kg, SD = 7.4). Subjects received
complimentary day-passes to a sports complex as an
incentive for their participation.
subject in a standard 12-lead recording arrangement. The
bipolar lead-II records gave the highest and clearest Twaves, therefore, they were selected for output channel
one. The signals from the pulse meter were recorded from
output channel two, and the aVf unipolar lead record was
read from channel three. The chart paper speed was set to
25 mm/s and the recorder sensitivity was set to 10
mm/mV.
3.1.2. Materials
A three-output channel computer software driven
Marquette electrocardiogram (Model Max-1) was
programmed for physiological recordings. Signals from a
photoplethysmograph (optical-sensor pulse meter-Model
CT-4600; IBS Co.) were fed into one of the input
channels of the electrocardiogram and thus these signals
were simultaneously recorded with the ECG. Marquette
Ag/AgCl electrodes (with pre-applied ECG electrolyte
paste) were used.
Cycling was performed on a Monark (Model 668)
stationary bicycle. A revolution counter (Lafayette
Instruments, Model 54419-A) was used to monitor
cycling pace. Habituation to the cycling rhythm was aided
with a Taktell Junior (Model 826) Metronome and a
digital RPM meter.
The MA was identical to that used in Experiment I, but
at the end three 7-point rating scales, ranging from 0 (not
at all) to 6 (extremely), were employed to assess the
subjectively perceived: (1) difficulty of each problem-set;
(2) effort exerted by the subjects to do well on each
problem-set; (3) importance to the subjects to perform well
on each problem-set (self-challenge), and (4) state of
relaxation while performing the MA task. All of these
measures were taken 1 min after the performance of the
MA task in each of the five conditions.
3.1.4. Procedure
The rest of the design and procedure (except for the
recording apparatus) was based on a previous study by
Szabo et al. (1994). In short, subjects were asked to only
have a light meal 2-3 h before the testing, and to refrain
from physical activity and caffeinated and alcoholic
beverages for 24 h prior to testing. At their arrival to the
laboratory, subjects were informed about the protocol.
Upon obtaining written consent for participation, subjects
were asked to complete the PAR-Q and the Exercise
Behavior Questionnaire. Next the electrodes were placed
on the subject's body followed by l-2 trial recordings. A
10 min rest interval, spent in an armchair, was
implemented to obtain a baseline recording necessary for
the calculation of the target HRs during cycling.
Subsequently, directions for the MA problems were given
along with a trial problem to ensure that the participant
had understood the directions. Five minutes later, the
first MA task was presented to habituate the subject to the
task. Performance was not recorded during the habituation
trial. Five minutes later testing began with the preexercise MA task that was presented for 1 min, while the
subject was still seated in the armchair. Subsequently, the
subject was led to the bicycle and instructed to cycle at 60
RPM in unison with the metronome click. Exercise
workload was progressively increased and by the end of
the first 5 min of the cycling a steady state in HR had been
reached, corresponding to 40% estimated MHRR. In the
10th min of the cycling, another 1 min MA task was
presented. Subsequently, the cycling pace was maintained
for another 5 mitt, then the workload was increased. Steady
state was reached again within the first 5 min of the
cycling that corresponded to 60% MHRR. In the 10th min
of cycling at 60% MHRR (25th min of cycling), another
MA task was presented and cycling continued at the same
intensity for another minute to obtain a
3.1.3. Electrocardiographic recordings
The photoplethysmograph transducer was placed on
the subjects' right index finger and secured in place with
a tape. Subjects were also instructed to minimize right arm
and hand movement during the test session. The
placement of the electrodes was preceded by hair shaving
if necessary, the light abrasion of the subject's skin until
it changed to a pinkish color, and the thorough washing of
the skin with 70% alcohol. Once the skin had dried completely, the prepared electrodes were placed on the
178
J.J. Furedy et al./Internationol Journal of Psychophysiology 22 (1996) 173-183
post-task ECG. Subsequently, the exercise intensity was
progressively decreased within a 3 min cool-down
period. Following exercise, the subject was seated in the
armchair and presented with two other MA tasks, 2 and 20
min, respectively, into the post-exercise recovery period.
The ECG recordings were obtained during the 10 s
before the presentation of each MA task, the last 10 s
during the minute of each MA task, and the last 10 s of
the minute following the termination of each task.
Subjective ratings of self-perceived difficulty, effort,
importance to the self, and state of relaxation were taken
verbally immediately after the last (post-MA) ECG
recording. Performances (ratio of answers to right answers)
were assessed later from the taped-record.
3.1.5. Data analysis
Heart rate and TWA were quantified in the same way as
in Experiment I. Pulse transit time was computed by
measuring the distance between the peak of the ECG Rwave
and
the
midpoint
of
the
subsequent
photoplethysmograph deflection. Quantifications were
performed by an independent reader who was unaware of
the various experimental conditions. All three measures
were averaged across the 10 s recording intervals.
Statistical procedures were carried out with the SYSTAT
computer software (Wilkinson, 1989).
A condition (5: pre-exercise, 40% MHRR exercise,
60% MHRR exercise, 2-min post-exercise and 20-min
post-exercise) by period (3: pre-, MA, and post-MA)
repeated measure ANOVA was used in testing the three
dependent variables. Period effects were of key interest in
the various conditions. Condition effects, if present, were
tested in limited context only to assess differences between
phases of: (1) low intensity cycling and rest, (2) low and
moderate intensity cycling, (3) rest and 20-min postexercise, and/or (4) 2- and 20-min post-exercise.
Effect sizes (Cohen, 1969) were calculated to assess
the size of changes from pre-MA to MA phase jointly
with the calculation of the percent changes. These
measures were deemed appropriate for comparing the
sensitivities of the three dependent measures to the
psychological challenge in the various experimental
conditions.
Fig. 3. Heart rate in five experimental conditions (rest, 40%
MHRR cycling, 60% MHRR cycling, 2-min post-exercise and
20-min post-exercise) and three sampling periods (pre-MA, during- MA, and post-MA) in each.
3.2. Results
3.2.1. Heart rate
The ANOVA for HR yielded a significant main effect
for condition (F (4,68) = 498, p< 0.001; GreenhouseGeisser (G-G) p < 0.001 E = 0.6748), a significant main
effect for period (F (2,34) = 34.3, p < 0.001; G-G p <
0.001, E = 0.5310), and a significant condition by period
interaction (F (8,136) = 3.0, p < 0.005; G-G p < 0.02, E =
0.6018). In testing the period main effect, simple contrasts
showed that in comparison to pre-MA, HR was greater
during MA in each of the five conditions ( p < 0.005;
Fig. 3). The condition effects were all significant (p <
0.001). Compared to rest, HR was still higher 20-min postexercise ( p < 0.001). The analysis of the interaction
confirmed that HRs in all the three phases were lower at
rest than 20-min post-exercise ( p < 0.001, 0.05. and
0.001, respectively). Further testing of the interaction was
not done, because most effects were 'induced' by the
combination of physical-mental challenges (for example it
was anticipated that during pre-MA or MA phase HR
would be lower at rest than during cycling at either
intensity or 2-min post-exercise).
J.J. Furedy et al./Intemational Journal of Psychophysiology 22 (1996) 173-183
3.2.2. T-wave amplitude
The ANOVA for TWA yielded a significant main
effect for condition (F (4,68) = 7.2, p < 0.001; G-G p<
0.001, E = 0.6304), a main effect for period (F (2,34) =
30.1, p< 0.001; G-G p< 0.001, E = 0.8518), and a
condition by period interaction (F &136) = 8.0, p<
0.001; G-G p< 0.001, E = 0.4949). Follow-up of the
period main effect showed significant differences
between TWA at pre-math and MA phase in all the five
conditions (p < 0.05). During both low and moderate
intensity exercise TWA increased from pre-MA to postMA (p < 0.005) and decreased from pre-MA to post-MA
at 2-min post-exercise (p < 0.005; Fig. 4).
The condition effects were due to lower TWA during
40% MHRR cycling than at rest (p < O.OOS), than at 60%
MHRR cycling (p < 0.01) and than at 20-min postexercise (p < 0.001). At 2-min post-exercise TWA was
greater than during 60% MHRR (p < 0.05). Condition
effects were not significant when contrasting rest and 20min post-exercise, 60% MHRR cycling and rest, 60%
MHRR cycling and 20-min post-exercise, and 2- and 20min post-exercise.
The follow-up of the interaction revealed that pre-MA
TWA was lower during 40% MHRR cycling than
during all other conditions (p < 0.05).
179
Fig. 5. Pulse-transit time in five experimental conditions (rest,
40% MHRR cycling, 60% MHRR cycling, 2-min post-exercise
and 20-min post-exercise) and three sampling periods (pre-MA,
during-MA, and post-MA) in each.
These effects were parallel during MA (p < 0.05) and
post-MA (p<0.05). Except at 40% MHRR cycling,
TWAs during stress were not different between the other
conditions. In comparison to rest, pre-MA TWA was
higher at both 2- and 20-min post-exercise.
Fig. 4. T-wave amplitude in five experimental conditions (rest,
40% MHRR cycling, 60% MHRR cycling, 2-min post-exercise
and 20-min post-exercise) and three sampling periods (pre-MA,
during-MA, and post-MA) in each.
3.2.3. Pulse-transit time
The ANOVA for PTT yielded a condition main effect
(F (4,68) = 18.8, p < 0.001; G-G p < 0.001, e =
0.4747), no period main effect, but a condition by period
interaction (F (8,136) = 3.0, p < 0.003; G-G p < 0.02, E
= 0.4942). Follow-up of the condition main effect showed
that during the 60% MHRR exercise PTT was lower than
in all the other conditions (p 0.001). The PTT in the 2-min
post-exercise condition was lower than at rest (p <
0.002) and than at 20-min post-exercise (p < 0.001). Other
differences were not significant (Fig. 5).
The analysis of the interaction showed that during preMA, MA and post-MA, PTT at 60% MHRR cycling was
lower than in all the other conditions (p < 0.001).
During MA at 40% MHRR cycling and 2-min postexercise, PTT was lower than at rest (p<0.05) and than at
20-min post-exercise ( p < 0.05). Finally, pre-MA PTT
was lower at 2-min post-exercise than at rest and than at 20min post-exercise ( p < 0.01).
180
J.J. Furedy et al./Internationol Journal of Psychophysiology 22 (1996) 173-183
Table I
Effect sizes (ES) and percent changes (change) in heartrate, T-wave amplitude, and pulse-transit time from premath to math phase in the five conditions (rest, 40 and
60% maximal reserve heart rate cycling, and 2 and 20
min after cycling)
Measure
Condition
ES
Heart rate
Rest
40% Exercise
60% Exercise
2-min after
20.min after
-1.03 L
-0.84 L
-0.71 M
-0.62 M
- 1.05 L
T-wave amplitude Rest
40% Exercise
60% Exercise
2-min after
20-min after
0.12 s
0.13 s
0.07 s
0.37 s
0.31 s
Pulse-transit time Rest
-0.03 s
40% Exercise
0.21 s
60% Exercise
0.37 s
2-min after
-0.51 M
20.min after
-0.01 s
3.2.5. Subjective measures
None of the subjective measures were different in the
five conditions except for relaxation (F(4,68) = 4.8, p<
0.002; G-G p< 0.004, E = 0.8002). Subjects reported
greater relaxation before and after exercise than during
exercise (p < 0.05).
Change (c/c)
18.8*
9.3 6.0*
9.6*
16.0 *
- 4.1 *
-6.7 *
-4 .1 *
-17.7 *
- 12.6 *
0.2
-2 .0
-3 .0
2.3
0.2
Note: An ES of around 0.2 reflects small (S) differences, an ES of
0.5 reflects moderate (M) differences, while an ES of 0.8 or
above reflects large (L) differences (Cohen, 1969); * indicates
that the changes from pre-math to math phase were statistically
significant in the corresponding condition.
3.2.4. Percent changes
Two other analysis were considered pertinent: (I) the
gauging of the change scores (or reactivity) to the
physiological challenge from rest to 40% MHRR cycling
and from the latter to 60% MHRR cycling, and (2) the
assessment of the changes in response to the psychological
challenge in the five conditions. In the former, percent
changes were calculated using the pre-MA records.
Mean percent values were 81 and 23% for HR, - 15 and
15% for TWA, and -2 and - 10% for PTT. For HR, the
change between rest and 40% MHRR cycling was greater
than that between the two exercise conditions (t (17)=
11.6, p < 0.001). For TWA the difference was also significant (t (17) = -3.0, p <0.01); and for PTT, the
difference between the two exercise loads significantly
exceeded that between the rest and the moderate-exercise
condition (t (17) = 2.2, p < 0.05). In the second, both
percent changes and effect sizes were calculated between
pre-MA and MA for each of the three dependent variables
(Table 1).
3.3. Discussion
Of the physiological measures, HR acceleration yielded
the clearest reactive-sensitivity results (see Fig. 3). It
increased reliably in response to the MA in all conditions
and in response to the physiological challenge of cycling
as well. In the latter case the increase was greater during
the moderate than during the light intensity exercise. The
HR recovery appeared to be relatively rapid in onset,
with most of the recovery occurring between the moderate
exercise challenge and 2-min post-exercise. However, 20
min after the exercise, HR was still greater than during
rest, indicating that total recovery from exercise did not
take place. In terms of effect sizes (ES) and percent
changes in response to MA, HR changes were moderate to
high (Cohen, 1969) and the largest among the three
dependent measures (see Table 1).
Although yielding ESs (Table 1) that were lower than
those for HR, TWA was sensitive to MA, because it
attenuated in all the five conditions relative to the
physiological state of rest and exercise. This effect may
be viewed as more potent than merely indicated by the
results, because in some instance TWA showed a
bidirectional response (i.e. increasing to tonic exercise and
decreasing to MA). Possibly, the augmentative effect of
exercise, that appeared after the 40% MHRR cycling, may
have partially cancelled out the attenuative effect of the
psychological challenge.
Indeed, TWA reaction to the physiological challenge
was more complex than to the psychological challenge.
Although the expected attenuation to the light exercise did
occur, and augmentation from this point was apparent that
persisted until 2-min post-exercise. In examining the preMA TWA values, it was apparent that a 'recovery' in TWA
has taken place between 2- and 20-min post-exercise as
showed by the lower TWA 20-min post-exercise than 2min post-exercise. Yet, 20-min post-exercise the TWA
was still higher, during pre-MA, than at rest. The
J.J. Furedy et al./Intemational Journal of Psychophysiology 22 (1996) 173-183
results not only show a reliable paradoxical augmentation
effect, but also indicate that this effect is not clearly timelocked to the condition (i.e. moderate exercise), but
appears to be maximal some 2 min following the
termination of cycling (refer to Fig. 4). The temporal
course of the TWA response to sustained exercise, as well
as its bidirectional reaction, suggests the operation of a
some homeostatic hormonal mechanism, perhaps
mediated by an increase in K + concentration (see e.g.
Blomquist, 1969; but for contrary evidence, see von
Duvillard et al., 19931, or other biochemical changes in
the face of the metabolic challenge posed by sustained
exercise.
With regard to physiological challenge, contractilitybased PTT yielded results that were at least as orderly as
those of HR, although the ESs of the supra-ventricular
measure were higher than those of either PTT or TWA.
Nevertheless, in terms of reactive sensitivity to the
physiological challenge, the PTT results were more
orderly than those of HR (refer to pre-MA values in Fig.
51, in the sense that recovery from exercise was complete
in PTT by 20 min following moderate exercise. Also,
whereas the greater increase in HR acceleration was from
rest to light (40% MHRR) exercise, PTT showed the reverse trend, with the greater attenuation occurring from
40% MHRR cycling to 60% MHRR cycling. This
HR/PTT difference is fitting the assumption that the
parasympathetically-influenced
atria1
measure
is
predominantly influenced by parasympathetic withdrawal
that occurs with even moderate exercise, whereas the
predominantly sympathetic ventricular measure reacts
more to harder exercise, when significant sympathetic
excitation occurs. It will be noted that without the lightmoderate exercise manipulation this differential HR/PTT
pattern, whatever its explanation, could not have been
seen.
On the other hand, and in contrast to TWA, the
ventricular PTT did not fare well as regards reactive
sensitivity to the psychological MA challenge. Particularly
discouraging was the fact that in the first, armchair-rest,
condition (which is the standard arrangement for assessing
psychological challenges) even the trend was not
suggestive of an attenuation to the arithmetic task. The
attenuative trend did emerge during the light-exercise
condition and persisted during the moderate-exercise
condition, but never reached significance. These results
parallel the
181
data obtained by Szabo (19931 who found that PTT did
not decrease in response to the same mental challenge
during armchair-rest but it decreased significantly during
orthostatic stress (standing). It appears that PTT responds
to this form of psychological challenge only when
another, most likely physical, type of challenge is also
present.
Finally, the behavioral measures assessed in Experiment
II
did
not
yield
additional
information
to
psychophysiological measures. The subjects did not
perceive the MA task more challenging while cycling
than during rest. This perceived task-consistency may be
taken as an assurance that changes observed in
psychophysiological measures were not related to certain
inconsistency in task demand characteristics. Subjects
reported greater state of relaxation during rest than during
exercise, indicating that they had associated this behavioral
measure with the physical state (i.e. exercise versus rest).
3.4. Conclusions
Concerning reactive sensitivity to psychological
challenge, both Experiments, as well as the literature,
suggest that atria1 HR acceleration is superior to either
PTT or TWA. As to the comparison PTT and TWA,
experiment II indicated that with the MA task used here,
TWA performed better than PTT. Consistent with our
results are those of Linden and Estrin (19881, where both
PTT and TWA were assessed in reaction to a somewhat
stronger MA task (solutions of mathematical equations
every 10 s). These investigators reported significant TWA
attenuation, but no reliable PTT change, to their task.
As to reactive sensitivity to the physical challenge the
'pattern' of differences seen in PTT (presumably more
reflective of sympathetic beta-adrenergic influence) were
superior than those seen in HR. For example, PTT was
more sensitive to the difference between light and
moderate exercise than to the difference between rest and
light exercise. On the other hand, TWA manifested the
paradoxical augmentation effect to moderate sustained
exercise, though not to light tonic exercise or to light
phasic exercise. Future factorial studies are needed, that
systematically vary both time course (phasic versus tonic)
and intensity of exercise, to determine whether the TWA
augmentation response (which may be
182
J.J. Furedy et al./Internationol Journal of Psychophysiology 22 (1996) 173-183
hormonally based) is not only a result of the nature
(physiological versus psychological), but also a result of
the time course (tonic versus phasic) of the challenge.
Taken together, the hereby reported results indicate
that: (1) HR is the most sensitive to both psychological
and physiological challenge, (2) TWA is sensitive to
psychological challenge, (3) TWA amplitude decreases in
response to brief (phas-ic)/light exercise and in response
to a short duration (about 10 min) tonic/light exercise, but
increases in response to longer duration (about 25 min)
tonic/moderate exercise, (4) the TWA augmentative
response is not limited to the instant of the 'higher
intensity' challenge, because it persisted at 2- and even at
20-min post-exercise, suggesting the involvement of
hormonal regulatory mechanisms, (5) PTT is not sensitive
to MA in resting position, but shows the tendency to
decrease during exercise, and (6) PTT is sensitive to
changes in the degree of physical challenge.
Acknowledgements
The authors wish to thank Andrea Szalai for her help
in data compilation, Lise Gauvin and John Spence for
their comments on various aspects of the study, Brian
Jarvis for his assistance in file transmission between two
universities, and Arthur Long and Daniel Desrochers for
their technical assistance. Special gratitude is extended to
two anonymous reviewers for their constructive
comments.
References
Furedy, J.J. (1987) Beyond heart rate in the cardiac psychophysiological assessment of mental effort: The T-wave amplitude
component of the electrocardiogram. Hum. Factors, 29:
183-194.
Furedy, J.J. and Heslegrave, R.J. (1984). The relative
sensitivities of heart rate and T-wave amplitude to stress:
Comments on, and some alternative interpretations of,
Penzien et al.'s results. Biol. Psychol., 19: 55-61.
Furedy, J.J., Heslegrave, R.J. and Scher, H. (1992) T-wave amplitude utility revisited: Some physiological and psychophysio
logical considerations. Biol. Psychol., 33: 241-248.
Godin, G., Jobin, J. and Bouillon, J. (1986) Assessment of leisure
time exercise behavior by self-report: A concurrent validity
study. Can. J. Publ. Health, 77: 359-362.
Hartung, G.H. (1972) Exercise electrocardiography in athletes and
nontrained subjects. J. Sports Med. Phys. Fitness, 12:
1866192.
Hartung, G.H. and Nouri S. (1979) The precordial T-wave during
exercise and recovery in middle aged runners and non-exercisers. J. Sports Med., 19: 285-290.
Heslegrave, R.J. and Furedy, J.J. (1980) Carotid dP/dt as a
psychophysiological index of sympathetic myocardial
effects: Some considerations. Psychophysiology, 17: 482494.
Hijzen, T.H., Van Der Gugten, J. and Bouter, L. (1984) Active
and passive coping under different degrees of stress:
Effects on urinary and plasma catecholamines and ECG Twave. Biol. Psychol., 18: 23-32.
Linden, W. and Estrin, R. (1988) Computerized cardiovascular
monitoring: Method and data. Psychophysiology, 25: 227234.
Blomquist, C.G. (1969) Variations of electrocardiographic response to exercise in different experimental conditions; Deconditioning, reconditioning and high altitude. In H.
Blackbum
(Ed),
Measurement
in
Exercise
Electrocardiography, Charles C. Thomas, Springfield IL.,
pp. 323-341.
Mayhew, J.L. (1971) Effect of endurance training on the Twave of the electrocardiogram of adult men. Med. Sci.
Sports Exert., 3: 172-174.
British Columbia Ministry of Health (19781 Physical Activity
Readiness Questionnaire (PAR-Q) Validation Report.
Roth, D.L., Bachtler, S.D. and Fillingim, R.B. (1990) Acute
emotional and cardiovascular effects of stressful mental
work during aerobic exercise. Psychophysiology, 27: 694701.
Brown, T.G., Szabo, A. and Seraganian, P. (1988) Physical
versus psychological determinants of heart rate reactivity to
mental arithmetic. Psychophysiology, 25: 532-537.
Cohen, J. (1969) Statistical Power Analysis for the Behavioral
Sciences, (2nd ed.), Academic Press, New York.
Contrada, R.J. (1992) T-wave amplitude: On the meaning of a
psychophysiological index. Biol. Psychol., 33: 249-258.
Contrada, R.J., Dimsdale, J., Levy, L. and Weiss, T. (1991)
Effects of isoproterenol on T-wave amplitude and heart rate:
A dose-response study. Psychophysiology, 28: 458-462.
Contrada, R.J., Krantz, D.S., Durel, L.A., Levy, L., LaRiccia, P.J.,
Anderson, J.R. and Weiss, T. (1989) Effects of betaadren-ergic
activity
on
T-wave
amplitude.
Psychophysiology, 26: 488-492.
Rose, K.D., Dunn, F.L. and Bargen, D. (1966) Serum electrolyte
relationship to electrocardiographic change in exercising a
athletes. J. Am. Med. Assoc., 195: 111-114.
Scher, H., Furedy, J.J. and Heslegrave, R.J. (1984) Phasic T-wave
amplitude and heart rate changes as indices of mental
effort and task incentive. Psychophysiology, 21: 326-333.
Szabo, A. (1993) The combined effects of orthostatic and mental
stress on heart rate, T-wave amplitude, and pulse transit
time. Eur. J. Appl. Physiol. Occup. Physiol., 67: 540-544.
Szabo, A. and Gauvin, L. (1992a) Mathematical performance
before, during, and following cycling at workloads of low
and moderate intensity. Percept. Mot. Skills, 75: 915-918.
Szabo, A. and Gauvin, L. (1992bl Reactivity to written mental
arithmetic: Effects of exercise lay-off and habituation,
Physiol. Behav., 51: 501-506.
J.J. Furedy et al./Intemational Journal of Psychophysiology 22 (1996) 173-183
Szabo, A., Peronnet, F., Gamin, L. and Furedy, J.J.
(1994) Mental challenge elicits 'additional'
increases in heart rate during low and moderate
intensity cycling. Int. J. Psychophysiol., 17: 197-204.
von Duvillard, S.P., LeMura, L.M., Szmedra, L. and De
Vito, P. (1993) Interrelationship between plasma
potassium concentra-
183
tion, pulmonary ventilation and electrocardiographic
change during and after highly intense exercise. J.
Manipul. Physiol. Ther., 16: 238-244.
Wilkinson, L. (1989). SYSTAT: The system for statistics.
Systat Inc., Evanston.