ORIGINAL ARTICLE

Pedro Montoya · Stuart Brody · Katja Beck · Ralf VeitHarald Rau

Differential b- and a-adrenergic activation during psychological stress

Accepted: 3 September 1996

Abstract The responsivity of several cardiovascular in-dices to a computerized mental arithmetic stress and acold pressor stress were investigated in 22 healthy adultsubjects. The major findings were that the largely b-adrenergically driven T-wave amplitude, pre-ejectionperiod, R-wave to pulse interval, and left ventricularejection time values responded only to mental ar-ithmetic; a significant decrease in cardiac output andincrease in peripheral resistance were elicited during thecold pressor test; inter-beat-interval and subjective stressratings responded significantly to both stresses com-pared to baseline levels, but more intensely to mentalarithmetic than the cold pressor test; blood pressure,stroke volume and the maximum of the first derivativeof the raw impedance signal responded unspecifically toboth stresses. These findings support the idea that car-diovascular responses to psychological challenge dependon the level of cognitive processing required for the task.In addition, the superfluity of multiple variable mea-surements to study cardiovascular reactivity in such si-tuations is discussed.

Key words T-wave amplitude · Beta-adrenergicactivation · Impedance cardiography

Introduction

Cardiovascular reactivity to laboratory stresses has beendiscussed as one factor related to cardiovascular disease(e.g. Ditto 1993; Fredrikson and Matthews 1990; Turner1994). Heart rate and blood pressure responses to well-defined stresses are the classical indices of cardiovascularreactivity. However, these responses result from dualautonomic and other influences, while less summative

signs of cardiovascular functioning, such as T-waveamplitude (TWA) and contractility-based estimates [pre-ejection period (PEP), R-to pulse interval (RPI), andrate of change of pressure with time (dP/dt)] have beenshown to have largely sympathetic b-adrenergic inputs(e.g. Allen et al. 1987; Newlin and Levenson 1979; Obristet al. 1978, 1979, 1987; Schwartz and Weiss 1983;Sherwood et al. 1990). For instance, decreases in TWAhave been shown to be substantially attenuated bypharmacological b-blockade (Rau 1991), leading to theinference that TWA reactivity, and, to a lesser degree,heart rate or blood pressure reactivity are under primarysympathetic b-adrenergic control, at least in this situa-tion.

However, there is still a controversy concerning thepurity of sympathetic b-adrenergic influences on each ofthese measurements (e.g. Bernston et al. 1994; Contrada1992; Furedy and Heslegrave 1983; Furedy et al. 1984,1992; Furedy and Scher 1985; Heslegrave and Furedy1983; Schwartz and Weiss 1983). There are, for example,many other factors such as preload and afterload effects,which has been suggested may potentially obscure therelationship between contractility-based measurementsand b-adrenergic influences (Heslegrave and Furedy1983; Newlin and Levenson 1979). To examine sympa-thetic influence on cardiovascular reactivity duringstress, noninvasively measured impedance cardiographicbased measurements have also been used. Cardiac out-put and total peripheral resistance have served as esti-mates of sympathetic adrenergic activation of the heartand vasculature. An increase in total peripheral re-sistance accompanied by a decrease in cardiac outputhas been interpreted as indicating a-adrenergic dom-inance, while a decrease in total peripheral resistanceand an increase in cardiac output implies b-adrenergicdominance (for review see Turner 1994). Furthermore,b-receptor blockade has been shown to abolish the in-creased cardiac and the decreased vascular responses,while a-receptor blockade has abolished the decreasedcardiac and the increased vascular responses (Girdleret al. 1993). It has been suggested that changes in stroke

Eur J Appl Physiol (1997) 75: 256 – 262 Springer-Verlag 1997

P. Montoya (&) · S. Brody · K. Beck · R. Veit · H. RauInstitute of Medical Psychology and Behavioural NeurobiologyEberhard-Karls University, Gartenstr. 29, D-72074 Tubingen,Germany

volume and contractility estimates (such as the HeatherIndex) might be determined among others factors bypreload (according to Starling’s law), afterload, andheart rate (Papillo and Shapiro 1990). That is, the reg-ulation of the volume of blood ejected by the heartduring each beat should be the product of the interac-tion of sympathetic activation and cardiodynamic andhaemodynamic factors. The dz/dtmax represents themaximum of the first derivative of the raw impedancesignal, and has been used to compute stroke volumetogether with other cardiovascular parameters. In thecurrent study, we examined its usefulness [based upon itsanalogy to the rate of change of pressure with time in thecarotid artery (dP/dT)] as an estimate of cardiovascularreactivity.

Among the several tasks used to induce psychologicalchallenges in the laboratory, mental arithmetic and thecold-water stress have been described as perhaps themost well-known psychological stresses (Turner 1994),which have been found to increase heart rate and bloodpressure. The main purpose of this study was to in-vestigate the discriminative value of several cardiovas-cular indices and subjective stress ratings in respondingto these two psychological stresses. Considering the well-known influence of stress on cardiovascular functioningwe assumed:

1. That all estimates will respond significantly to bothstressors compared to resting, and

2. That only some of those estimates (the purer ones)will distinguish between the stresses,

while indices with mixed a-and b-adrenergic as well asparasympathetic influences (such as systolic and dia-stolic blood pressure, and inter-beat-interval) and sub-jective stress reports would respond unspecifically to thestressors. In keeping with earlier findings (Rau 1991), itwas also expected that the more specific estimates of b-adrenergic sympathetic myocardial activation such asresponses of the electrocardiographic TWA, RPI, andPEP suppression (relative to baseline) would occur to agreater degree during the computerized mental arith-metic task than during the cold pressor test, providing amore specific index of mental stress (Furedy et al. 1992).Changes in cardiac output and total peripheral re-sistance as well as stroke volume, dz/dt and myocardialcontractility as estimated by Heather index changes werealso analysed.

Methods

Subjects

A group of 25 healthy volunteer university students (13 men and 12women aged 19 to 28 (mean 22.7) years, served as subjects. Missingvalues in three of the subjects led to the results from 22 subjectsbeing analysed. All the subjects were paid for their participation (20DM plus a 0–10 DM bonus contingent on performance in thecomputer mental arithmetic task).

Procedure

The subjects were seated upright in a comfortable chair in a quietlaboratory and told that they could discontinue participation atany time. A 5-min computerized mental arithmetic task and a3-min hand cold pressor test were administered in counterbalancedorder with a 5-min rest preceding each task. At the end of thesession, data from a further 5-min relaxation were obtained andused in the present analyses as the post baseline reference. Aftereach recording period, subjective verbal stress ratings were ob-tained. The ratings consisted of the subjects reporting how stressedthey felt on a 10-point numerical scale (1 = minimum stress,10 = maximum stress).

Cold pressor test

The subjects were required to place their right hands approximately30-cm deep in a bucket filled with ice-water (4°C). They were in-structed to leave their hands there for 3 min. However, 2 subjectsremoved their hands after less than a minute and were excludedfrom further analyses for this reason.

Computerized mental arithmetic task

This task consisted of time-limited arithmetic problems (e.g.(10+8)/3; (20–12)/4) presented on the video monitor of an IBMcompatible computer. The subjects were required to enter the so-lution to the problem using the computer keyboard. Each problemwas followed immediately by the next one. Problems were accom-panied by a computer generated noise of increasing intensity andfrequency (until the problem was solved). The subjects were givenfeedback on the correctness of their response after each trial. Inaddition, one point was added to or subtracted from an initialbonus (30 points). The number of points achieved was alwaysdisplayed in the upper left corner of the screen. To motivate thesubjects, they were told that points could be exchanged for cash atthe end of the experiment (4 points = 1 DM).

Physiological recordings

Blood pressure was obtained with the FINAPRES device, whichemploys the vascular unloading principle (Rau et al. 1993; Wes-seling et al. 1986) by recording changes in photo-plethysmographvalues. The cuff of the FINAPRES was attached to the third fingerof the nondominant hand. Impedance cardiographic signals (dz/dtand z0) and electrocardiogram were obtained using a Diefenbachcardiodynograph with four adhesive band electrodes. Recordingelectrodes were placed around the base of the neck and the lowerthorax at the level of the xiphisternal joint. The current electrodeswere separated from the upper and lower recording electrodes byapproximately 3 and 5 cm, respectively. A current of 4 mA at100 kHz was passed through the two outer electrodes. Phono-cardiogram was obtained using a heart-sounds microphone placedon the left apex. All acquired analogue data (blood pressure curve,dz/dt, z0, electrocardiogram, and phonocardiogram) were digitizedby an A-D converter (DT2821) at 500 Hz and stored for later off-line processing. Data acquisition and data analysis programs werewritten in ASYST routines running on an AT-386 computer.

Data reduction and analysis

To enhance signal to noise ratio and for a better identification ofcomponents of the impedance cardiographic signal (dz/dt(max), B-,and X-point, all physiological signals were ensemble averaged off-line for three 1-min periods using the R-wave as trigger (accordingto the suggestion of Sherwood et al. 1990). The 3 min of the coldpressor test, and the 2nd, 3rd and 4th min of the mental arithmetictask were separately averaged and used for statistical analysis.

257

The following estimates were derived from each 1-min grandaverage:

A. Stroke volume (millilitres) was calculated according to Kubi-cek’s formula (c.f. Sherwood et al. 1990):

stroke volume = rho × (L/z0)2 × LVET × dz/dtmax

where L is the distance between the recording electrodes measuredat the front (in centimetres), z0 is the baseline impedance betweenthe recording electrodes (in ohm), LVET is left ventricular ejectiontime (in milliseconds), and dz/dtmax the maximum (ohm per second)of the dz/dt signal; rho represents the resistance of blood, and inthis case a value of 135 Ω × cm was used.

B. PEP was defined as the time interval between the occurrence ofthe electrocardiographic R-wave and the dz/dt B-wave (in milliseconds).

C. LVET was defined as the time interval between the occurrance ofthe B-wave and the X-point of the dz/dt signal (in milliseconds).

D. Cardiac output was estimated as the product of heart rate timesstroke volume (litres per minute).

E. Myocardial contractility (estimated as the Heather Index) wasdefined as dz/dtmax divided by the time interval between theelectrocardiogram Q-wave and the occurrence of dz/dtmax (ohmper second squared).

F. Systolic and diastolic blood pressures were determined by themaximal and minimal values of the averaged blood pressurecurve (in millimetres of mercury) (Wolk and Velden 1991).

G. Total peripheral resistance was defined as the ratio betweenmean blood pressure and cardiac output multiplied by 80 (dyne-seconds per centimetre to the power of 5).

H. RPI was measured as the delay of the systolic blood pressurepeak from the R-wave of the electrocardiogram.

I. TWA was defined as the maximum of the electrocardiogramwithin a window of 100–300 ms after the R-wave of the elec-trocardiogram, referred to the level of the isoelectric PQ-inter-val which was measured as the average of the milliseconds50–40 before the occurrence of the R-wave (Rau 1991).

Individualized period z-scores were computed for each index bycomputing means and standard deviations for task and rest periodsand subtracting this task and rest period mean from the value of agiven period and dividing by the standard deviation. These z-scoreswere entered into a one-way ANOVA with repeated measures(Greenhouse-Geisser adjusted degrees of freedom) to test the in-fluence of task (cold pressor vs mental arithmetic) on stress ratingsand each cardiovascular index.

Results

Figure 1 displays the reactivity of each estimate, allow-ing some comparison of the magnitudes of the indices.

The F-values in Table 1 reflect the overall ANOVAfor each estimate, and these are arranged according tothe similarity of their responses. All estimates were sig-nificantly affected by the factor task. Table 1 also givesthe t-values for task-induced changes (referred to thepreceding baseline) and for the difference between taskvalues.

As depicted in Figure 1, subjective stress ratings andthe 12 psychophysiological estimates may be condensedinto five groups on the basis of their responses to stress.Interbeat interval and subjective stress displayed an or-dinal trend, as mental arithmetic had a greater influencethan the cold pressor test, which was more stressful thanbaseline. Both blood pressures, stroke volume, and dz/dtwere influenced in either stress condition relative tobaseline, but with no difference between the stresses. The

TWA, PEP, RPI, and LVET were affected only bymental arithmetic relative to baseline. Cardiac outputand peripheral resistance were reactive to the coldpressor test but not to mental arithmetic. The Heatherindex responded differentially, yielding lower values thanbaseline for the cold pressor test, and higher values thanbaseline for mental arithmetic.

Discussion

In the present study, several cardiovascular responsesand stress ratings to two well-known psychologicalchallenges were investigated. Cardiovascular indices in-cluded the classical ones (heart rate, blood pressure), aswell as others obtained from noninvasive impedancecardiography. The results indicated

1. That stresses elicit significant changes in cardiovas-cular functioning and strain ratings compared toresting levels, and

2. That specific cardiovascular response patterns areelicited by each psychological stress.

As expected, all indices (cardiovascular and sub-jective stress ratings) showed significant changes in atleast one stress condition compared to the rest periods.Consistent with our earlier results (Brody et al. 1994,1996), we found that the report of subjective stress wasthe most reactive index. Among the physiological in-dices, systolic blood pressure was the most reactive tothe cold pressor stress test, and RPI was the most re-active to mental arithmetic compared to baseline. Thus,investigators seeking to minimize b error might chooseindices that are more reactive (assuming that a less re-active index does not possess features that are otherwisedesirable for the investigation in hand, such as the re-lative ease of measurement of TWA).

We also found that some indices responded morespecifically than others to the stresses. On the basis oftheir reactivity pattern, indices were arranged in fivegroups: specifically reactive to mental arithmetic, speci-fically reactive to the cold pressor test, more reactive tomental arithmetic than to the cold pressor test, differ-ently reactive to both stressors, and nonspecifically re-active. In the first group, cardiovascular indices such asTWA, PEP, LVET, and RPI responded significantly tomental arithmetic compared to resting levels. All theseindices showed no significant response to the coldpressor stress. Prior research has suggested that cogni-tive stress induces primarily b-adrenergic activity (Obrist1981). Results of many other studies have indicated thatPEP (Allen et al. 1987; Bernston et al. 1994; Cacioppoet al. 1994; Light and Obrist 1983), RPI (Obrist et al.1979), and TWA (Contrada et al. 1989; Contrada 1992;Furedy and Heslegrave 1983; Furedy et al. 1984, 1992;Furedy and Scher 1985; Heslegrave and Furedy, 1979,1983; Rau 1991) might be considered as noninvasiveindices of b-adrenergic influence on the heart. Bothfindings are supported by the present results which, in

258

addition to many other papers, clustered TWA togetherwith the relatively purely sympathetically driven PEP,and RPI. The present results also verify TWA as anindicator of b-adrenergic influence on sympathetic

myocardial activity. Thus, given the relative simplicity ofthe recording and analysis of the electrocardiographsignal, measurement of TWA might have some ad-vantages over impedance cardiograph and blood pres-sure (RPI) related indices for evaluating sympatheticinfluences on myocardial activity.

The second group of indices (total peripheral re-sistance and cardiac output) was reactive to the coldpressor test, but not to mental arithmetic. It is of interest

Fig. 1 Means and standard error bars of within subjects z-scores foreach of the indices investigated. The indices have been arranged in fivegroups according to the type of reactivity observed (see Table 1)

259

that these cardiovascular indices are those that havebeen considered as the haemodynamic determinants ofblood pressure changes (Sherwood and Turner 1993;Turner 1994). Considering that significant increases inblood pressure were elicited to a similar degree duringboth tasks (Fig. 1), it is noteworthy that our data in-dicated that the factors underlying these changes (totalperipheral resistance and cardiac output) were quitedifferent during the two stresses. During the cold pressortask compared to the rest period, significantly increasedtotal peripheral resistance and decreased cardiac outputwere found. This finding is in accordance with data fromother studies (e.g. Allen and Crowell 1989; Allen et al.1992; Turner 1994), suggesting that an increased totalperipheral resistance response could be related to adominant a-adrenergic activation. However, it is possi-ble that the cold pressor task has other, possibly con-founding effects as has been suggested by Peckermanet al. (1994). Firstly, the local physical demands thatcold places on thermoregulation may produce vaso-motor changes, which may be reflected in changes inperipheral resistance and blood pressure. Secondly, coldis not only a source of stress, it is also a source of pain,which also implies that the stress experienced in the coldpressor test is to some degree pain as well. As Turner(1994) has noted, one may cope with the stress of thecold pressor test by employing defocusing strategies.Thus, cold pressor tasks would involve pain and localthermoregulatory processes, as well as cognitive factors.Finer cardiovascular psychophysiological differentiation

of mental activity would involve an absence of per-turbing physical challenges, and greater variability ofmental tasks. Nevertheless, a very important differencebetween the two stresses used in the present experimentwas the necessary cognitive effort to cope with thechallenge1. Mental arithmetic requires considerablymore cognitive processing to reach a solution than thecold pressor task, where in the response to which acognitive processing of the stimuli does not seem to behelpful. Future research should evaluate b-adrenergicactivation as a function of the level of cognitive pro-cessing required by the stressful situation.

During the mental arithmetic stress, significant in-crease compared to the rest period of blood pressure wereachieved without changes in total peripheral resistanceand cardiac output. Considering that mental arithmeticwas characterized by strong b-adrenergic activation, wewould suggest that sympathetic activity was also re-sponsible for the blood pressure increase during themental arithmetic task. A question remains as to whycardiac output was not affected if increased b-adrenergicactivation was present. One possible mechanism might bethat increased blood pressure impeded the ejection ofblood from the left ventricle into the circulatory system(afterload effect), resulting in a reduction in stroke vo-lume, and therefore, in cardiac output. Thus, it seems

1 We thank an anonymous reviewer for the helpful comment on therole of the level of congnitive processing in cardiovascular reactivity

Table 1 Values for overall F from the ANOVA test, t for com-parisons of responses to the cold pressor test and mental arithmeticto baseline values and comparisons of the responses produced by

the two stresses for each of the indices investigated. TWA T-waveamplitude, PEP pre-ejection period, LVET left ventricular ejectiontime, RPI R-to pulse interval

Indices Overall F D ColdPressortest (t)

D Mentalarithmetic(t)

Mentalarithmeticvs coldpressor test (t)

Computer task reactiveTWA 3.0 * 1.0 2.1 * 3.1 **

PEP 27.3 *** 0.3 8.2 *** 8.1 ***

LVET 3.4 * 0.3 2.6 * 2.3 *

RPI 45.2 *** 0.7 11.0 *** 9.9 ***

Cold pressor reactiveTotal peripheral resistance 11.4 *** 6.0 *** 1.8 3.4 **

Cardiac output 3.7 * 2.6 * 1.6 3.8 **

Differentially reactiveMyocardial contractility 6.2 ** 3.1 ** 2.4 * 4.7 ***

More reactive to computer task thanto cold pressorStress rating 83.4 *** 9.1 *** 13.0 *** 4.5 ***

Inter-beat interval 35.9*** 2.3 * 8.7*** 6.1 ***

Nonspecifically reactiveSystolic blood pressure 47.7 *** 8.9 *** 10.0 *** 1.5Diastolic blood pressure 38.0 *** 8.1 *** 8.0 *** 0.01Stroke volume 5.8 ** 2.9 ** 2.6 * 0.4dz/dt 7.2 ** 4.5 *** 2.5 * 0.96

*P < 0.05, **P < 0.01, ***P < 0.001

260

that under strong b-adrenergic activation blood pressurechanges might occur without the influence of increasedcardiac output.

In the third group we included two indices (interbeatinterval and subjective stress rating) which respondedsemispecifically to the stressors, in that the computertask produced a more intense reaction than the coldpressor test, which still had a significant effect. Heartrate and the reporting of subjective stress were sig-nificantly increased during the cold pressor test andmental arithmetic compared to baseline values. How-ever, the fact that a significantly greater increase wasfound during mental arithmetic than during the coldpressor test led us to assign interbeat interval and sub-jective stress to a separate group of response patterns.The interbeat interval result was consistent with that ofFahrenberg et al. (1993). As for subjective stress, thebroadness of the index was likely to have included bothunpleasant effort and physical discomfort. On the otherhand, it is possible that with a larger, or a differentsample of subjects, the fine line separating this group ofresponse patterns from the nonspecifically reactivegroup might be blurred.

Myocardial contractility (as indicated by the Heatherindex) showed a differential response pattern, and wasincluded in the fourth group. For this index, a significantincrease was found during mental arithmetic, whereas asignificant decrease was elicited during the cold pressortest. Increased contractility has been considered to re-flect b-adrenergic activity (Kelsey and Guethlein 1990).Significant decreases in myocardial contractility duringthe cold pressor task have also been reported (Allen et al.1992; Peckerman et al. 1994; Wilson et al. 1991) andinterpreted as reflecting an attenuation of b-adrenergicinvolvement. The current results further confirm thatdistinct autonomic patterns are present in the mentalarithmetic (b-adrenergic) and cold pressor (a-adrenergic)tasks.

Finally, in the fifth group there were some indices(blood pressure, stroke volume and dz/dt) which did notdifferentiate significantly between the stresses. Duringboth stresses compared to baseline levels, significant in-creases were found in systolic and diastolic blood pres-sures, and decreases in stroke volume and dz/dtmax. Theblood pressure results are not surprising considering thata-adrenergic as well as b-adrenergic activation have beenshown to lead to an increased blood pressure response(e.g. Turner 1994). Regarding stroke volume and dz/dt,our results supported the view that these indices are un-der the control of mixed cardiovascular influences such aspreload and afterload, as well as sympathetic activation.

Thus, the results of the present experiment indicatedthat mental arithmetic elicited an enhanced b-adrenergicactivation response characterized by significant de-creases in TWA, PEP, RPI, LVET, interbeat interval,stroke volume, and dz/dt together with significant in-creases in blood pressure and myocardial contractilitycompared to baseline levels. On the other hand, the coldpressor test, evoked significant increases in total

peripheral resistance and blood pressure together withsignificant decreases in cardiac output, interbeat inter-val, myocardial contractility, stroke volume, and dz/dt.These responses should be considered signs of a-adre-nergic activation. Thus, our findings would suggest adifferent effect of psychological stress on cardiovascularfunctioning depending on the level of cognitive proces-sing required by the tasks.

An implicit purpose of our approach was to de-termine if there was any clear merit in employing somany dependent variables to evaluate cardiovascularreactivity to stress. Similar reaction among two or moreof the variables would imply that (at least for such si-tuations), some indices are superfluous.

References

Allen MT, Crowell MD (1989) Pattern of autonomic responseduring laboratory stressors. Psychophysiology 26:603–614

Allen MT, Obrist PA, Sherwood A, Crowell MD (1987) Evaluationof myocardial and peripheral vascular responses during reac-tion time, mental arithmetic and cold pressor tasks. Psycho-physiology 24:648–656

Allen MT, Shelley KS, Boquet AJ (1992) A comparison of cardi-ovascular and autonomic adjustments to three types of coldstimulation tasks. Int J Psychophysiol 13:59–69

Bernston GG, Cacioppo JT, Binkley PF, Uchino BN, Quigley KS,Fieldstone A (1994) Autonomic cardiac control. III. Psycholo-gical stress and cardiac response in autonomic space as revealedby pharmacological blockades. Psychophysiology 31:599–608

Brody S, Maier C, Montoya P, Rau H (1994) Speaking from theheart: cardiovascular components of stress rating changes andthe relative reactivity of physiological and psychological vari-ables. Eur J Appl Physiol 69:277–280

Brody S, Veit R, Rau H (1996) Neuroticism but not cardiovascularstress reactivity is associated with less longitudinal blood pres-sure increase. Person Ind Diff 20:375–380

Cacioppo JT, Uchino BN, Bernston GG (1994) Individual differ-ences in the autonomic origins of heart rate reactivity: thepsychometrics of respiratory sinus arrhythmia and preejectionperiod. Psychophysiology 31:412–419

Contrada RJ (1992) T-wave amplitude: on the meaning of a psy-chophysiological index. Biol Psychol 33:249–258

Contrada RJ, Krantz DW, Durel LA, Levy L, LaRiccia PL, An-derson JR, Weiss T (1989) Effects of beta-adrenergic activity onT-wave amplitude. Psychophysiology 26:488–492

Ditto B (1993) Familial influences on heart rate, blood pressure,and self-report anxiety responses to stress: results from 100 twinpairs. Psychophysiology 30:635–645

Fahrenberg J, Foerster F, Wilmers F (1993) Cardiovascular re-sponse to mental and physical tasks as predictors of ambulatorymeasurements. J Psychophysiol 7:275–289

Fredrikson M, Matthews KA (1990) Cardiovascular responses tobehavioral stress and hypertension: a meta-analytic review. AnnBehav Med 12:30–39

Furedy JJ, Heslegrave RJ (1983) A consideration of recent criti-cisms of the T-Wave amplitude index of myocardial sympa-thetic activity. Psychophysiology 20:204–211

Furedy JJ, Scher H (1985) Intuitive and factual approaches in thecomparison of contractility and repolarization measures ofsympathetic myocardial activity. Int J Psychophysiol 2:235–237

Furedy JJ, Heslegrave RJ, Scher H (1984) Psychophysiological andphysiological aspects of T-wave amplitude in the objectivestudy of behavior. Pavlov J Biol Sci 19:182–194

Furedy JJ, Heslegrave RJ, Scher H (1992) T-wave amplitude utilityrevisited: some physiological and psychophysiological con-siderations. Biol Psychol 33:241–248

261

Girdler SS, Hinderliter AL, Light KC (1993) Peripheral adrenergicreceptor contributions to cardiovascular reactivity: Influence ofrace and gender. J Psychosom Res 37:177–193

Heslegrave RJ, Furedy JJ (1979) Sensitivities of HR and T-waveamplitude for detecting cognitive and anticipatory stress. Phy-siol Behav 22:17–23

Heslegrave RJ, Furedy JJ (1983) On the utility of T-wave amplitude:a reply to Schwartz and Weiss. Psychophysiology 20:702–707

Kelsey RM, Guethlein W (1990) An evaluation of the ensembleaveraged impedance cardiogram. Psychophysiology 27:24–33

Light KC, Obrist PA (1983) Task difficulty, heart rate reactivityand cardiovascular responses to an appetitive reaction timetask. Psychophysiology 20:301–312

Newlin DB, Levenson RW (1979) Pre-ejection period: measuringbeta-adrenergic influences upon the heart. Psychophysiology16:546–553

Obrist PA (1981) Cardiovascular psychophysiology: a perspective.Plenum Press, New York

Obrist PA, Gabelein CJ, Teller ES, Langer AW, Grignolo A, LightKC, McCubbin JA (1978) The relationship among heart rate,carotid dp/dt and blood pressure in humans as a function oftype of stress. Psychophysiology 15:102–115

Obrist PA, Light KC, McCubbin JA, Hutcheson JS, Hoffer JL(1979) Pulse transit time: relationship to blood pressure andmyocardial performance. Psychophysiology 16:292–301

Obrist PA, Light KC, James SA, Strogatz DS (1987) Cardiovas-cular responses to stress. I. Measures of myocardial responseand relationship to high resting systolic pressure and parentalhypertension. Psychophysiology 24:65–78

Papillo JF, Shapiro D (1990) The cardiovascular system. In: Ca-cioppo JT, Tassinary LG (eds) Principles of psychophysiology.Cambridge University Press, Cambridge, pp 456–512

Peckerman A, Hurwitz BE, Saab PG, Llabre MM, McCabe PM,Schneiderman N (1994) Stimulus dimensions of the cold pressor

test and the associated patterns of cardiovascular response.Psychophysiology 31:282–290

Rau H (1991) Responses of the T-wave amplitude as a function ofactive and passive tasks and beta-adrenergic blockade. Psy-chophysiology 28:231–239

Rau H, Brody S, Droste C, Kardos A (1993) Blood pressurechanges validate phase related external suction, a controlledmethod for stimulation of human baroreceptors. Eur J ApplPhysiol 67:26–29

Schwartz PH, Weiss T (1983) T-wave amplitude as an index ofcardiac sympathetic activity: a misleading concept. Psycho-physiology 20:696–701

Sherwood A, Turner JR (1993) Postural stability of hemodynamicresponses during mental challenge. Psychophysiology 30:237–244

Sherwood A, Allen MT, Fahrenberg J, Kelsey RM, Lovallo WR,Doornen JP van (1990) Methodological guidelines for im-pedance cardiography. Psychophysiology 27:1–23

Turner JR (1994) Cardiovascular reactivity and stress. PlenumPress, New York

Wesseling KH, Settels JJ, deWit B (1986) The measurement ofcontinuous finger arterial pressure noninvasively in stationarysubjects. In: Schmidt TH, Dembroski TM, Blumchen G (eds)Biological and psychological factors in cardiovascular disease.Springer, Berlin Heidelberg, New York, pp 355–375

Wilson BL, Albright GL, Steiner SS, Andreassi JL (1991) Cardi-odynamic response to psychological and cold pressor stress:further evidence for stimulus response specificity and direc-tional fractionation. Biofeed Self Regul 16:45–53

Wolk C, Velden M (1991) Measurement of phasic blood pressurechanges with the FINAPRES-System. In:Ruddel H, Curion I(eds) Non-invasive continuous blood pressure measurements.Peter Lang, Frankfurt, pp 61–78

262