The heart contracts to reward: Monetary incentives and

preejection period

MICHAEL RICHTER and GUIDO H. E. GENDOLLASection of Psychology, University of Geneva, Geneva, Switzerland

Abstract

Wright’s (1996) integration of motivational intensity theory (Brehm & Self, 1989) and Obrist’s (1981) active coping

approach predict that cardiovascular reactivity in active coping depends on the importance of success when task

difficulty is unclear. Despite the support for this perspective, one of the basic hypothesesFthe mediation of these

effects by beta-adrenergic activityFhas not been tested yet. To close this gap, participants worked on a delayed-

matching-to-sample task and could earn either 1, 15, or 30 Swiss Francs for a successful performance. Results showed

that preejection period reactivityFan indicator of beta-adrenergic impact on the heartFincreased with increasing

incentive value. Thus, this experiment closes a gap in the support of Wright’s model by demonstrating that beta-

adrenergic reactivity is associated with incentive value under conditions of unclear difficulty.

Descriptors: Cardiovascular reactivity, Incentive, Reward, Beta-adrenergic activity, Active coping

According tomotivational intensity theory (Brehm&Self, 1989),

task difficulty and the importance of success determine energy

mobilization in instrumental tasks (i.e., tasks that allow attaining

a personal goal). Wright (1996) adapted these predictions to

psychophysiology by integrating themwithObrist’s (1981) active

coping approach. The resulting integrative model deals with car-

diovascular adjustments in active coping tasks and is based on

the following two central predictions. (1) If task difficulty is

known, cardiovascular reactivity should be proportional to task

difficulty as long as success is possible and as long as the nec-

essary energy is justified by the importance of success. (2) If no

information is available about task difficulty (i.e., when task

difficulty is unclear), the importance of task success should di-

rectly determine cardiovascular reactivity until performers have

attained their maximum effort level. Once an individual’s max-

imum level has been reached, further increases in importance

should not make a difference. Numerous studies investigated

these hypotheses in the last 20 years and showed that adjustments

in blood pressure and heart rate (HR) follow the integrative

model’s predictions. Especially systolic blood pressure (SBP) re-

liably reacted to manipulations of task difficulty and success im-

portance (for reviews, seeGendolla&Brinkmann, 2005;Gendolla

& Wright, 2005; Richter, Gendolla, & Krusken, 2006; Wright,

1996, 1998; Wright & Kirby, 2001).

Despite this interest in the predictions of Wright’s model, re-

search that investigated the physiological mechanisms underly-

ing the observed cardiovascular adjustments is rare. According

to Wright (1996), energy mobilization in active coping should

be mediated by sympathetic (beta-adrenergic) impact on the

heart. Correspondingly, effects of task difficulty and success

importance on cardiovascular parameters should be due to

changes in beta-adrenergic activity. It follows that cardiovascular

measures that strongly depend on the force of myocardial con-

tractionFwhich is determined by beta-adrenergic dischargeFshould be more strongly influenced by task difficulty and success

importance than measures that are less strongly associated with

the force of contraction. This basic assumption of the integrative

model has often been used to explain the fact that SBP showed

the predicted effects more consistently than HR, diastolic blood

pressure (DBP), or mean arterial blood pressure (MAP; e.g.,

Gendolla &Wright, 2005, for a review). SBP strongly depends on

the force of myocardial contraction, whereas HR, DBP, and

MAP are less systematically associated with the force of myo-

cardial contraction (e.g., Berne & Levy, 1977; Levick, 2003).

However, increases in blood pressure are not an un-

ambiguous indicator of increases in myocardial contractility or

beta-adrenergic activity, respectively. Increases in peripheral re-

sistanceFsympathetically or parasympathetically mediatedFmay have similar effects on blood pressure as increases in

myocardial beta-adrenergic activity. Furthermore, the effects of

increased myocardial contractility on blood pressure can be coun-

teracted by accompanying dilative effects on the vasculature.

Although beta-adrenergic activity has been a topic of discus-

sion in the research onWright’s integrative approach, researchers

only recently started to directly address this issue by assessing

preejection period (PEP). PEPFthe time interval between the

We thank Kerstin Brinkmann for helpful comments on an early draft

of this article. The preparation of this article was facilitated by a research

grant of the Swiss National Science Foundation awarded to the first

author.Address reprint requests to: Michael Richter, University of Geneva,

FPSE, Department of Psychology, 40, Bd. du Pont-d’Arve, CH-1211Geneva 4, Switzerland. E-mail: Michael.Richter@unige.ch

Psychophysiology, 46 (2009), 451–457. Wiley Periodicals, Inc. Printed in the USA.Copyright r 2009 Society for Psychophysiological ResearchDOI: 10.1111/j.1469-8986.2009.00795.x

451

onset of ventricular depolarization and the opening of the aortic

valveFis a valid and reliable indicator of beta-adrenergic influ-

ences on the heart (e.g., Benschop et al., 1994; Harris,

Schoenfeld, &Weissler, 1967; Newlin & Levenson, 1979; Obrist,

Light, James, & Strogatz, 1987; Schachinger, Weinbacher, Kiss,

Ritz, & Langewitz, 2001). Annis, Wright, and Williams (2001)

manipulated subjective task difficulty by presenting different

difficulty levels of an active coping task together with a

bogus ability feedback. They observed that PEP reactivity

increased with increasing task difficulty as long as the task was

not too difficult. However, PEP reactivity scores were in

general slightly positive, indicating that participants invested

less energy during task performance than during habituation.

Thus, these data provide only weak support for Wright’s

hypotheses concerning the mobilization of energy during task

performance.

Using a more straightforward experimental design, Richter,

Friedrich, and Gendolla (2008) manipulated the objective diffi-

culty of a memory task and assessed PEP reactivity. Their results

were in accordance with the predictions of the integrative model.

When task success was possible, changes in PEP were propor-

tional to increases in task difficulty. When task difficulty was so

high that success was impossible, PEP reactivity was low. More-

over, there are several studies conducted outside the frame of

Wright’s integrative model that support the association between

PEP and task difficulty, as well (Kelsey, 1991; Light & Obrist,

1983; Sherwood, Davis, Dolan, & Light, 1992; Tomaka & Pala-

cios-Esquivel, 1997). However, other studies did not find any

association between task difficulty and PEP (Harrell & Clark,

1985; Kelsey et al., 1999; Sherwood, Royal, & Light, 1993).

Nevertheless, the studies that found a reliable difficulty–PEP as-

sociation consistently showed the pattern predicted by Wright’s

integrative model and, therefore, support its predictions about

the influence of task difficulty on cardiac reactivity when infor-

mation about task difficulty is available and clear.

However, evidence is still lacking for cardiovascular adjust-

ments under conditions of unclear task difficulty. Recently,

Richter and Gendolla (2006, 2007) demonstrated that the value

of rewardFone variable that determines success impor-

tanceFinfluences SBP reactivity when task difficulty is unclear.

They found SBP increases with reward under these conditions.

However, as in most of the preceding research, the authors only

assessed blood pressure and HR but not more valid indices of

beta-adrenergic influences on the heart. Therefore, these studies

cannot be considered as unambiguous support for the assump-

tion that beta-adrenergic activity underlies cardiovascular reac-

tivity in active coping with unclear task difficulty.

The present experiment aims to close this gap by investigating

the influence of incentive value on PEP reactivity. For this pur-

pose, participants performed a delayed-matching-to-sample task

with unclear task difficulty under one of threemonetary incentive

conditions (1 Swiss Franc vs. 15 Swiss Francs vs. 30 Swiss

Francs). Based on Wright’s integrative model, we expected that

PEP reactivity would proportionally increase with increasing in-

centive value: PEP reactivity should be low in the 1-Swiss-Franc

condition, moderate in the 15-Swiss-Francs condition, and high

in the 30-Swiss-Francs condition. Because preceding research

has reliably demonstrated that SBP reactivity follows the pre-

dictions of Wright’s approach (e.g., Gendolla & Wright, 2005,

for a review) we predicted the same pattern for systolic reactivity.

We made no specific predictions for the reactivity of HR, DBP,

and MAP because these parameters are only loosely associated

with beta-adrenergic influences on the myocardium (e.g., Berne

& Levy, 1977; Levick, 2003).

Method

Participants and Design

Thirty-one psychology students (mean age 28 years) participated

for course credit. They were randomly assigned to one of three

experimental conditions (incentive: 1 Swiss Franc vs. 15 Swiss

Francs vs. 30 Swiss Francs). The distributions of women and

men were balanced (9 women and 2 men in the 1-Swiss-Franc

cell, 8 women and 2 men in both the 15-Swiss-Francs and the

30-Swiss-Francs cells). Participation was anonymous and

voluntary.

Apparatus and Physiological Measurement

Cardiovascular measures were assessed during two periods: ha-

bituation and task performance. A Vasotrac APM205A system

(Medwave, Arden Hills, MN) measured SBP, DBP, and MAP

(all measures in millimeters of mercury [mmHg]). The system’s

cuff was placed around the wrist of the participant’s nondom-

inant arm and one blood pressuremeasure was obtained every 12

to 15 heart beats. Electrocardiogram (ECG) and thoracic im-

pedance (impedance cardiogram, ICG) signals were sampled

with 800 Hz using a Cardioscreen 1000 system (medis, Ilmenau,

Germany; see Scherhag, Kaden, Kentschke, Sueselbeck, &

Borggrefe, 2005, for a validation of the system). Four pairs of

disposable spot electrodes were placed on the right and left sides

of the base of the participant’s neck and on the left and right

middle axillary lines at the height of the xiphoid. All obtained

measures and signals were directly stored on a computer disk so

that participants and experimenter were ignorant of all values

obtained during the experiment. Experiment generation software

(INQUISIT by Millisecond Software, Seattle, WA) controlled

the presentation of stimuli and instructions as well as the collec-

tion of participants’ responses.

Procedure

The experiment was run in individual sessions. The experimenter

assessed participants’ weight and height and applied the blood

pressure cuff and the spot electrodes. Then, participants an-

swered some biographical questions and were instructed to relax

and to sit as quietly as possible during the next minutes of ha-

bituation. Blood pressure measures were taken in intervals of 12

to 15 heart beats during the following 10 min of habituation.

ECG and ICG were continuously measured. After the habitu-

ation period participants received the task instructions.

We used a slightly modified version of the delayed-matching-

to-sample task employed in an experiment by Thorne, Dawson,

and Schell (2006). The task consisted of 28 trials each lasting for a

total time of 11.5 s. At the beginning of each trial, a mask con-

sisting of 45 points painted in dark gray was presented. After 400

ms some of the points changed their color to light gray. This

patternFthe sample patternFlasted for 2400 ms. Then, the

mask was displayed again for a duration of 800 ms. It was fol-

lowed by a pattern consisting of either 2, 4, 6, 8, 10, or 12 points

painted in light gray. This target pattern was masked after 1600

ms by a slide asking participants to decide if the pattern of light

gray points differed between sample and target pattern by press-

ing either one of two keys. In half of the trials the target pattern

showed the same pattern of light gray points as the sample pat-

452 M. Richter and G.H.E. Gendolla

tern. In the other half of the trials the position of one point

differed between both. The number of light gray points did al-

ways match. Different patterns of 2, 4, 10, and 12 light gray

points were presented four times during each task; patterns con-

sisting of 6 and 8 points were presented three times. The sequence

of the trials was randomized. After having pressed a key, par-

ticipants received feedback that their response had been re-

corded. Independent of response time, the software always

waited for the total of 11.5 s before advancing to the next trial. If

participants did not respond within this time frame the software

automatically advanced to the next trial.

Corresponding to the manipulations that have been previ-

ously employed to create tasks with unclear difficulty (Richter &

Gendolla, 2006, 2007), task instructions explained only the gen-

eral procedure of the task. Task instructions did not inform par-

ticipants about the number of light gray points, presentations

times, or total performance time. Furthermore, participants

learned that they could win a prize for successful task perfor-

mance. They were also informed that the computer would ran-

domly assign to them a performance standard at the end of the

task. If their performance matched this standard, they would

receive a monetary reward. Participants could win either 1 Swiss

Franc (about USD 0.85), 15 Swiss Francs (about USD 12.50), or

30 Swiss Francs (about USD 25.00). It was emphasized that the

standard could range anywhere between 1 and 28 correct trials.

Thus, the performance standard or task difficulty, respectively,

was unclear to participants.

Next, participants rated the importance of task success

(‘‘How important does task success appear to you?’’) on a scale

ranging from 1 (unimportant) to 9 (very important) and incentive

attractiveness (‘‘How attractive does winning the reward appear

to you?’’) on a scale ranging from 1 (unattractive) to 9 (very

attractive). Participants then performed the 28 task trials. Blood

pressure measures were collected in intervals of 12 to 15 heart

beats; ECG and ICG were continuously assessed.1After task

performance participants received their reward if applicable.

They were debriefed, probed for suspicion, and received their

course credit.

Data Scoring, Reduction, and Analysis

ECG and ICG signals were automatically down-sampled to 200

Hz by the Cardioscreen 1000 system and were processed using a

software developed for our laboratory (Richter, 2006). A thresh-

old peak-detection algorithm automatically identified R-peaks,

which were visually confirmed afterward (ectopic beats were re-

placed using a nonlinear predictive interpolation; see Lippman,

Stein, & Lerman, 1994). HR (in beats per minutes) was calcu-

lated based on the detected R-peaks. The first derivative of the

change in thoracic impedance was calculated and the resulting

dZ/dt signal was ensemble averaged over periods of 20 s using the

detectedR-peaks (Kelsey&Guethlein, 1990; Kelsey et al., 1998).

Only artifact-free cycles were used to construct the ensemble

averages. R-onset and B-point were scored for each ensemble

average, visually inspected, and, if necessary, corrected as sug-

gested by Sherwood et al. (1990).2 PEP (in milliseconds) was

determined as the interval between R-onset and B-point (Bernt-

son, Lozano, Chen, & Cacioppo, 2004).

The arithmeticmeans of themeasures obtained during the last

4 min of habituation constituted our SBP, DBP, MAP, HR, and

PEP baseline scores (all Cronbach’s as4.98).3 The arithmetic

means of the measures assessed during the first 4 min of task

performance were used as SBP, DBP, MAP, HR, and PEP task

scores (all Cronbach’s as4.99). Cardiovascular reactivity

(change) scores were computed for each participant and each

cardiovascular parameter by subtracting baseline scores from

their respective task scores (Llabre, Spitzer, Saab, Ironson, &

Schneiderman, 1991). Reactivity scores were not significantly

correlated with their respective baseline scores (� .14ors

o� .04, ps4.48). Therefore, reactivity scores were not adjusted

with regard to baseline values (see Benjamin, 1967; Llabre et al.,

1991). Correlations between cardiovascular baseline and reac-

tivity measures can be found in Table 1. All cardiovascular re-

activity scores were analyzed using single factor ANOVAs.

To examine our theory-driven predictions about the impact of

incentive value on beta-adrenergic reactivity in more detail, we

compared the reactivity scores of PEP and SBP between the

1-Swiss-Franc group and the 15-Swiss-Franc group as well as

between the 15-Swiss-Franc group and the 30-Swiss-Franc

group using t tests. Given that our predictions were clearly di-

rectional, all t tests were tested one-tailed. Because beta-

adrenergic activity is not a major determinant of HR, DBP,

and MAP, we made no specific predictions for these measures

and did not apply the t tests to these measures.

Results

Preliminary Analyses

Statistical analyses of blood pressure values are based on 30

participants. Due to equipment malfunction 1 participant in the

1-Swiss-Franc condition had less than four blood pressure mea-

sures during baseline and task performance and was, therefore,

excluded from the analyses. Two (gender) � 3 (incentive value)

between-persons ANOVAs found significant effects of gender

on PEP baseline, F(1,25)5 18.43, po.001, Z2p ¼ :42, and on

DBP reactivity scores, F(1,24)5 4.28, po.05, Z2p ¼ :15. Fur-

thermore, the gender � incentive value interaction on HR base-

line scores was significant, F(2,25)5 5.13, po.02, Z2p ¼ :29.However, given the low number of male participants, these re-

sults should be regarded with caution. Instead of using gender as

a covariate in our analyses of cardiovascular measures, we re-

peated all analyses with a sample restricted to female partic-

ipants. Because the results of these analyses did not differ from

those conducted with the whole sample, we report in the follow-

ing only the results for the whole sample.

Based on the assessed weights and heights of the participants

we calculated participants’ body mass indexes (BMI). BMI val-

ues were significantly correlated with DBP (r5 � .39) andMAP

(r5 � .37) reactivity (pso.05). All other cardiovascular mea-

sures were not significantly associated with BMI values,

� .36orso.33, ps4.06. Because including BMI values in the

analyses of DBP and MAP reactivity resulted in a significant

Incentives and preejection period 453

1Due to the Vasotrac’s samplingmode the number of samples assesseddiffered with HR. Single factor analyses of variance (ANOVAs) of thevariance within baseline and task samples did not showa significant effectof incentive value on any of the three cardiovascular measures. Further-more, HR did not significantly correlate with the variance of SBP, MAP,or DBP samples. Thus, the different numbers of samples for each par-ticipant and period did not lead to differences in the stability of our bloodpressure measures.

2The software used the R-Z interval to estimate B-point location asdescribed in Lozano et al. (2007).

3Cronbach’s as for HR, SBP, DBP, andMAP scores are based on 1-min averages.

covariate effect, Fs(1,25)44.95, pso.04, we analyzed BMI-cor-

rected DBP and MAP reactivity.

Cardiovascular Baselines

Single factor between-persons ANOVAs did not show significant

baseline differences between the experimental conditions (all

ps4.28). Means and standard errors of all cardiovascular base-

line measures are displayed in Table 2.

Cardiovascular Reactivity

Preejection period reactivity. The single factor ANOVA

showed a significant effect of incentive value, F(2,28)5 8.91,

po.01,MSE5 14.35, Z2p ¼ :39. Both t tests were significant: The1-Swiss-Franc cell (M5 0.97, SE5 1.16) differed from the 15-

Swiss-Francs cell (M5 � 2.40, SE5 1.26), t(19)5 1.96, po.04,

and the 15-Swiss-Francs cell differed from the 30-Swiss-Francs

cell (M5 � 6.01, SE5 1.10), t(18)5 2.16, po.03. Figure 1 re-

flects that the pattern appeared as predicted.

Blood pressure reactivity. The single factor ANOVA resulted

in a significant effect for incentive value on SBP reactivity,

F(2,27)5 3.46, po.05, MSE5 72.97, Z2p ¼ :20. As predicted,

the 1-Swiss-Franc cell (M5 4.12, SE5 2.60) reliably differed

from the 15-Swiss-Franc cell (M5 12.55, SE5 3.28),

t(18)5 2.01, po.02. However, the difference between the 15-

Swiss-Francs cell and the 30-Swiss-Francs cell (M5 13.08,

SE5 2.09) was not significant, t(18)5 0.14, p4.45. Figure 2

shows this pattern of SBP reactivity. Single factor analyses of

covariance (ANCOVAs) showed an effect of incentive value nei-

ther on BMI-corrected DBP reactivity, F(2,25)5 2.19, p4.13,

MSE5 54.39, Z2p ¼ :15, nor on BMI-corrected MAP reactivity,

F(2,25)5 2.90, p4.07, MSE5 58.54, Z2p ¼ :19.4 Means and

standard errors of BMI-corrected DBP and MAP reactivity

scores appear in Table 3.

Heart rate reactivity. The single factorANOVAdid not show

a significant effect of incentive value, F(2,28)5 0.98, p4.38,

MSE5 7.67, Z2p ¼ :07. Means and standard errors of HR reac-

tivity scores appear in Table 3.

Task Ratings

Ratings of reward attractiveness and success importance were

significantly correlated, r5 .39, po.04, and a multivariate anal-

ysis of both measures found a significant incentive effect,

F(4,54)5 6.35, po.001, Z2p ¼ :32. Because both ratings should

indicate the effectiveness of our manipulation of incentive value,

we further analyzed them with the same t tests as PEP and SBP

reactivity. Attractiveness ratings significantly differed between

the 1-Swiss-Franc cell (M5 2.27, SE5 0.27) and the 15-Swiss-

Francs cell (M5 3.70,SE5 0.45), t(19)5 2.78, po.02, aswell as

between the 15-Swiss-Francs cell and the 30-Swiss-Francs cell

(M5 5.00, SE5 0.56), t(18)5 1.82, po.05. Importance ratings

only differed between the 1-Swiss-Franc cell (M5 3.00

SE5 0.45) and the 15-Swiss-Francs cell (M5 4.90, SE5 0.31),

t(19)5 3.41, po.01. The difference between the 15-Swiss-Franc

cell and the 30-Swiss-Francs cell (M5 4.50, SE5 0.43) was not

reliable, t(18)5 0.75, po.24. Reward attractiveness significantly

correlated with PEP reactivity, r5 � .52, po.01; all other cor-

relations between task ratings and reactivity scores were not sig-

nificant,� .19orso.33, ps4.07.

Task Performance

Single factor ANOVAs demonstrated that participants did not

differ in the percentage of correctly solved trials, F(2,28)5 0.79,

p4.46, MSE5 95.70, Z2p ¼ :05, or in their reaction times,

F(2,28)5 0.80, p4.46, MSE5 58,877, Z2p ¼ :05. Means and

standard errors of the performance measures appear in Table 4.

Mediation Analysis

We further examined the relationships between our incentive

manipulation, task ratings, and cardiovascular reactivity using

Sobel tests (Preacher & Hayes, 2004). None of the tests was

454 M. Richter and G.H.E. Gendolla

Table 1. Correlations between Cadiovascular Baselines and Reactivity Scores

Measures 1 2 3 4 5 6 7 8 9 10

1. PEP baseline F .22 .25 .32 .27 .09 .24 .22 .20 .262. SBP baseline F .99n .97n .15 .32 .12 .03 .03 .143. MAP baseline F .99n .18 .29 .13 .05 .04 .144. DBP baseline F .22 .25 .12 .05 .05 .145. HR baseline F .11 .27 .24 .26 .136. PEP reactivity F .35 .39n .41n .307. SBP reactivity F .98n .95n .088. MAP reactivity F .98n .079. DBP reactivity F .0310. HR reactivity F

n5 30 for all correlations including blood pressure measures, n5 31 for all other correlations. PEP: preejection period,SBP: systolic blood pressure, MAP: mean arterial blood pressure, DBP: diastolic blood pressure, HR: heart rate.npo.05.

Table 2. Cell Means and Standard Errors of Cardiovascular

Baseline Scores

Mean Standard error

1 SwissFranc

15 SwissFrancs

30 SwissFrancs

1 SwissFranc

15 SwissFrancs

30 SwissFrancs

PEP baseline 102.99 107.96 105.91 4.15 3.33 2.53HR baseline 75.99 76.83 75.04 2.79 3.60 3.52SBP baseline 125.01 116.74 129.35 5.83 2.80 7.26DBP baseline 67.31 62.59 69.50 4.68 1.92 5.50MAP baseline 88.74 82.34 92.55 5.32 2.44 6.44

n5 11 in the 1-Swiss-Franc cell of HR and PEP baseline, n5 10 in allother cells. PEP: preejection period, HR: heart rate, SBP: systolic bloodpressure, DBP: diastolic blood pressure, MAP: mean arterial bloodpressure. Preejection period is in milliseconds, heart rate is in beats perminute, and systolic blood pressure, diastolic blood pressure, and meanarterial blood pressure are in millimeters of mercury.

4Analyzing uncorrected DBP and MAP reactivity resulted in thesame pattern of results: Neither MAP nor DBP reactivity showed a sig-nificant influence of incentive value (ps4.10).

significant (ps4.14). However, examining the indirect effect by

bootstrapping (number of bootstrap resamples was 5,000)Fwhich is less conservative than the Sobel testFresulted in a

confidence interval that did not include zero for the indirect effect

of incentive value, task importance, and PEP reactivity,

M5 0.73, CI.95 5 0.02, 1.84. All other confidence intervals in-

cluded zero. In sum, these results suggest that the effect of in-

centive value on PEP reactivity was mediated by subjective task

importance.

Discussion

Wright’s (1996) integration of motivational intensity theory

(Brehm& Self, 1989) and Obrist’s (1981) active coping approach

posits that success importance determines cardiac activity during

active copingwith unclear task difficulty. PEP and SBP reactivity

data of our experiment corroborate this hypothesis. PEP reac-

tivity depended on the level of monetary incentiveFour manip-

ulation of success importanceFoffered for successful task

performance: PEP reactivity was low in the 1-Swiss-Franc con-

dition, moderate in the 15-Swiss-Francs condition, and high in

the 30-Swiss-Francs condition. SBP reactivity showed a similar

pattern. However, the difference between the 15-Swiss-Francs

and the 30-Swiss-Francs groups was not significant. DBP,MAP,

and HR reactivity were not reliably influenced by the manipu-

lation of incentive value.

The pattern of PEP reactivity clearly supports Wright’s no-

tion that incentive effects on energy mobilization under condi-

tions of unclear task difficulty are mediated by beta-adrenergic

impact on the heart. Because changes in PEP reliably reflect beta-

adrenergic activity on the heart under most conditions (e.g.,

Harris et al., 1967; Lewis, Rittogers, Forester, & Boudoulas,

1977; Sherwood et al., 1990), we interpret our findings as re-

flecting the expected association between incentive value and

beta-adrenergic activity. However, changes in ventricular filling

or aortic diastolic pressure can also result in changes of PEP

(e.g., Lewis, Leighton, Forester, & Weissler, 1974). For a valid

interpretation of PEP it is, therefore, recommended to evaluate

PEP reactivity only in the light of the reactivity of HR and DBP

(Obrist et al., 1987; Sherwood et al., 1990). If changes in PEP are

caused by changes in aortic diastolic pressure, one would expect

that increases in PEP reactivity would go along with decreases in

DBP. Increases in ventricular filling that cause increases in PEP

should be accompanied by decreases in HR (Obrist et al., 1987).

Thus, if changes in PEP are caused by changes in aortic diastolic

pressure and/or ventricular filling, DBP and/or HR should

decrease with increasing PEP. This was clearly not the case in our

study. Therefore, it seems unlikely that ventricular filling or aor-

tic diastolic pressure can account for the observed effects on PEP.

Our results are also congruent with preceding research on

cardiovascular reactivity in active coping with unclear difficulty

(e.g., Richter & Gendolla, 2006, 2007). Corresponding to the

results of these experiments, SBP reactivity showed an influence

of incentive value, as well. However, these effects were small

compared to incentive effects on PEP reactivity. The match be-

Incentives and preejection period 455

Figure 1. Cell means and standard errors of preejection period (PEP)

reactivity during task performance.

Figure 2.Cell means and standard errors of systolic blood pressure (SBP)

reactivity during task performance.

Table 3. Cell Means and Standard Errors of HR, DBP, and MAP

Reactivity Scores

Mean Standard error

1 SwissFranc

15 SwissFrancs

30 SwissFrancs

1 SwissFranc

15 SwissFrancs

30 SwissFrancs

HR reactivity 0.77 1.11 2.39 0.62 1.04 0.91DBP reactivity 2.74 7.64 9.68 2.46 2.33 2.33MAP reactivity 3.14 9.66 11.17 2.55 2.42 2.42

n5 11 in the 1-Swiss-Franc cell of HR reactivity, n5 10 in all other cells.HR: heart rate, DBP: diastolic blood pressure,MAP:mean arterial bloodpressure. Heart rate is in beats per minute and diastolic blood pressureand mean arterial blood pressure are in millimeters of mercury and cor-rected for the influences of BMI.

Table 4. Cell Means and Standard Errors of Performance Scores

Mean Standard Error

1 SwissFranc

15 SwissFrancs

30 SwissFrancs

1 SwissFranc

15 SwissFrancs

30 SwissFrancs

% correct 64.29 67.14 69.64 2.59 2.43 3.96Reaction time 860.02 946.74 991.01 65.76 96.37 61.20

n5 11 in the 1-Swiss-Franc cell, n5 10 in both other cells. % correct:percentage of correctly solved trials.

tween both cardiovascular measures suggests that SBP reactivity

effects were mainly driven by changes in the force of myocardial

contraction. Because incentive value should affect beta-ad-

renergic impact on the heart it is not surprising that DBP,

MAP, and HR only roughly resembled PEP reactivity and that

they were not reliably associated with incentive value. Moreover,

this pattern of results is congruent with preceding research on the

integrative model that has consistently found effects on SBP re-

activity but only rarely on other cardiovascular parameters (for

reviews, see, e.g., Gendolla & Wright, 2005; Richter et al., 2006;

Wright, 1996, 1998). Our results also replicate previous studies

by Wright and colleagues that investigated the influence of in-

centive value on cardiovascular reactivity (Wright, Killebrew, &

Pimpalapure, 2002; Wright, Tunstall, Williams, Goodwin, &

Harmon-Jones, 1995). They showed that systolic reactivity in-

creased with increasing incentive valueFmanipulated either by

social evaluation or by monetary rewardFwhen performers

could choose their own performance standard. DBP, MAP, and

HR reactivity were not systematically influenced by incentive

value.

Our results also relate to preceding research that has studied

the relationship between reward and cardiovascular responding.

Fowles and other researchers postulated a positive relationship

between reward value and heart rate and supported this hypoth-

esis with empirical data (Belanger & Feldman, 1962; Elliot, 1969;

Fowles, 1988; Fowles, Fisher, & Tranel, 1982; Tranel, 1983;

Tranel, Fisher, & Fowles, 1982). Our data replicate these studies

insofar as heart rate changes were the most pronounced at the

highest level of incentive value. However, the differences in heart

rate reactivity between the incentive groups were not so pro-

nounced that they would support the notion of a linear or a

monotonic relationship between incentive value and changes in

heart rate. Based on Fowles’s work and Gray’s motivation the-

ory (e.g., Gray, 1982), researchers have recently started to use

cardiac activity as an indicator of approach motivation or an

activation of the motivational approach system. For instance,

Brenner, Beauchaine, and Sylvers (2005) found that preejection

period and heart rate were sensitive to interindividual differences

in reward-related approach motivation. Our data suggest that

there is also a situational influence of reward value on cardiac

activity that should be considered when cardiac measures are

used as indicator of approach motivation.

In sum, our results clearly support the hypothesis of Wright’s

integrative approach that energy mobilization under conditions

of unclear task difficulty is related to incentive value. Moreover,

the observed association between incentive value and preejection

period reactivity suggests that incentive effects in active coping

with unclear difficulty are based on the effects of incentives on

beta-adrenergic activity on the heart. Thereby, our research clar-

ifies preceding studies that have relied on less direct measures of

beta-adrenergic activity and closes a gap in the empirical support

for Wright’s integrative model.

REFERENCES

Annis, S., Wright, R. A., &Williams, B. J. (2001). Interactional influenceof ability perception and task demand on cardiovascular response:Appetitive effects at three levels of challenge. Journal of Applied Bio-behavioral Research, 6, 82–107.

Belanger, D., & Feldman, S. M. (1962). Effects of water deprivationupon heart rate and instrumental activity in the rat. Journal of Com-parative and Physiological Psychology, 55, 220–225.

Benjamin, L. (1967). Facts and artifacts in using analysis of covariance to‘‘undo’’ the law of initial values. Psychophysiology, 4, 187–206.

Benschop, R. J., Nieuwenhuis, E. E. S., Tromp, E. A.M., Godaert, G. L.R., Ballieux, R. E., et al. (1994). Effects of �-adrenergic blockade onimmunologic and cardiovascular changes induced by mental stress.Circulation, 89, 762–769.

Berne, R. M., & Levy, M. N. (1977). Cardiovascular physiology ((3rded.)). St. Louis, MO: C. V. Mosby.

Berntson, G. G., Lozano, D. L., Chen, Y.-J., & Cacioppo, J. T. (2004).Where to Q in PEP. Psychophysiology, 41, 333–337.

Brehm, J. W., & Self, E. A. (1989). The intensity of motivation. AnnualReview of Psychology, 40, 109–131.

Brenner, S. L., Beauchaine, T. P., & Sylvers, P. D. (2005). A comparisonof psychophysiological and self-report measures of BAS and BIS ac-tivation. Psychophysiology, 42, 108–115.

Elliot, R. (1969). Tonic heart rate: Experiment on the effects of collativevariables lead to a hypothesis about its motivational significance.Journal of Personality and Social Psychology, 12, 211–228.

Fowles, D. C. (1988). Psychophysiology and psychopathology: A mo-tivational approach. Psychophysiology, 25, 373–391.

Fowles, D. C., Fisher, A. E., & Tranel, D. T. (1982). The heart beats toreward: The effect of monetary incentive on heart rate. Psychophys-iology, 19, 506–513.

Gendolla, G. H. E., & Brinkmann, K. (2005). The role of mood states inself-regulation: Effects on action preferences and resource mobiliza-tion. European Psychologist, 10, 187–198.

Gendolla, G.H. E., &Wright, R. A. (2005).Motivation in social settings:Studies of effort related cardiovascular arousal. In J. P. Forgas, K.Williams, & W. von Hippel (Eds.), Social motivation (pp. 205–1244).New York: Cambridge University Press.

Gray, J. A. (1982). Precis of the neuropsychology of anxiety: An enquiryinto the functions of the septo-hippocampal system. The Behavioraland Brain Sciences, 5, 469–534.

Harrell, J. P., & Clark, V. R. (1985). Cardiac responses to psychologicaltasks: Impedance cardiographic studies. Biological Psychology, 20,261–283.

Harris, W. S., Schoenfeld, C. D., & Weissler, A. M. (1967). Effects ofadrenergic receptor activation and blockade on the systolic preejec-tion period, heart rate, and arterial pressure in man. Journal of Clin-ical Investigation, 46, 1704–1714.

Kelsey, R. M. (1991). Electrodermal lability andmyocardial reactivity tostress. Psychophysiology, 28, 619–631.

Kelsey, R. M., Blascovich, J., Tomaka, J., Leitten, C. L., Schneider, T.R., & Wiens, S. (1999). Cardiovascular reactivity and adaptation torecurrent psychological stress: Effects of prior task exposure. Psy-chophysiology, 36, 818–831.

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

Kelsey, R. M., Reiff, S., Wiens, S., Schneider, T. R., Mezzacappa, E. S.,& Guethlein, W. (1998). The ensemble-averaged impedance cardio-gram: An evaluation of scoring methods and interrater reliability.Psychophysiology, 35, 337–340.

Levick, J. R. (2003). An introduction to cardiovascular physiology. Lon-don: Oxford University Press.

Lewis, R. P., Leighton, R. F., Forester, W. F., &Weissler, A. M. (1974).Systolic time intervals. In A. M. Weissler (Ed.), Noninvasive cardiol-ogy (pp. 301–368). New York: Grune & Stratton.

Lewis, R. P., Rittogers, S. E., Forester,W. F., &Boudoulas,H. (1977). Acritical review of the systolic time intervals. Circulation, 56, 146–158.

Light, K. C., & Obrist, P. A. (1983). Task difficulty, heart rate reactivity,and cardiovascular responses to an appetitive reaction time task.Psychophysiology, 20, 301–311.

Lippman, N., Stein, K. M., & Lerman, B. B. (1994). Comparison ofmethods for removal of ectopy in measurement of heart rate vari-ability. AJP – Heart and Circulatory Physiology, 267, 441–418.

Llabre, M. M., Spitzer, S. B., Saab, P. G., Ironson, G. H., & Schnei-derman, N. (1991). The reliability and specificity of delta versus re-

456 M. Richter and G.H.E. Gendolla

sidualized change as measure of cardiovascular reactivity to behav-ioral challenges. Psychophysiology, 28, 701–711.

Lozano, D. L., Norman, G., Knox, D., Wood, B. L., Miller, D., Emery,C. F., et al. (2007). Where to B in dZ/dt. Psychophysiology, 44, 113–119.

Newlin, D. B., & Levenson, R. W. (1979). Pre-ejection period: Measur-ing beta-adrenergic influences upon the heart. Psychophysiology, 16,546–553.

Obrist, P. A. (1981). Cardiovascular psychophysiology: A perspective.New York: Plenum.

Obrist, P. A., Light, K. C., James, S. A., & Strogatz, D. S. (1987).Cardiovascular responses to stress: I. Measures of myocardial re-sponse and relationships to high resting systolic pressure and parentalhypertension. Psychophysiology, 24, 65–78.

Preacher, K. J., & Hayes, A. (2004). SPSS and SAS procedures for es-timating indirect effects in simple mediation models. Behavior Re-search Methods, Instruments, & Computers, 36, 717–731.

Richter, M. (2006). BlueBox (version 1.11) [computer software].Richter, M., Friedrich, A., & Gendolla, G. H. E. (2008). Task difficulty

influences on cardiac reactivity. Psychophysiology, 45, 869–875.Richter, M., & Gendolla, G. H. E. (2006). Incentive effects on cardio-

vascular reactivity in active coping with unclear task difficulty. Inter-national Journal of Psychophysiology, 61, 216–225.

Richter, M., & Gendolla, G. H. E. (2007). Incentive value, unclear taskdifficulty, and cardiovascular reactivity in active coping. InternationalJournal of Psychophysiology, 63, 294–301.

Richter, M., Gendolla, G. H. E., & Krusken, J. (2006). Context-depen-dent mood effects on mental effort mobilization: A view from theMood-Behavior-Model. In A. V. Clark (Ed.), The psychology ofmoods (pp. 57–79). New York: Hauppauge.

Schachinger, H., Weinbacher, M., Kiss, A., Ritz, R., & Langewitz, W.(2001). Cardiovascular indices of peripheral and central sympatheticactivation. Psychosomatic Medicine, 63, 788–796.

Scherhag, A., Kaden, J. J., Kentschke, E., Sueselbeck, T., & Borggrefe,M. (2005). Comparison of impedance cardiography and thermodi-lution-derived measurements of stroke volume and cardiac output atrest and during exercise testing. Cardiovascular Drugs and Therapy,19, 141–147.

Sherwood, A., Allen, M. T., Fahrenberg, J., Kelsey, R. M., Lovallo, W.R., & van Doornen, L. J. P. (1990). Methodological guidelines forimpedance cardiography. Psychophysiology, 27, 1–23.

Sherwood, A., Davis, M. R., Dolan, C. A., & Light, K. C. (1992). Car-diovascular reactivity assessment: Effects of choice of difficulty onlaboratory task responses. International Journal of Psychophysiology,12, 87–94.

Sherwood, A., Royal, S. A., & Light, K. C. (1993). Laboratory reactivityassessment: Effects of casual blood pressure status and choice of taskdifficulty. International Journal of Psychophysiology, 14, 81–95.

Thorne, G. L., Dawson, M. E., & Schell, A. M. (2006). Effects of per-ceptual load on startle reflex modification at a long lead interval.Psychophysiology, 43, 498–503.

Tomaka, J., & Palacios-Esquivel, R. L. (1997).Motivational systems andstress-related cardiovascular reactivity. Motivation and Emotion, 21,275–295.

Tranel, D. T. (1983). The effects of monetary incentive and frustrativenon-reward on heart rate and electrodermal activity. Psychophysiol-ogy, 20, 652–657.

Tranel, D. T., Fisher, A. E., & Fowles, D. C. (1982). Magnitude ofincentive effects on heart rate. Psychophysiology, 19, 514–519.

Wright, R. A. (1996). Brehm’s theory of motivation as a model of effortand cardiovascular response. In P. M. Gollwitzer & J. A. Bargh(Eds.), The psychology of action: Linking cognition and motivation tobehavior (pp. 424–453). New York: Guilford.

Wright, R. A. (1998). Ability perception and cardiovascular response tobehavioral challenge. In M. Kofta, G. Weary, & G. Sedek (Eds.),Control in action: Cognitive and motivational mechanisms (pp. 197–232). New York: Plenum.

Wright, R. A., Killebrew, K., & Pimpalapure, D. (2002). Cardiovascularincentive effects where a challenge is unfixed: Demonstrations in-volving social evaluation, evaluator status, and monetary reward.Psychophysiology, 39, 188–197.

Wright, R. A., & Kirby, L. D. (2001). Effort determination of cardio-vascular response: An integrative analysis with applications in socialpsychology. In M. P. Zanna (Ed.), Advances in experimental socialpsychology (vol. 33, pp. 255–307). San Diego, CA: Academic Press.

Wright, R. A., Tunstall, A. M., Williams, B. J., Goodwin, J. S., & Har-mon-Jones, E. (1995). Social evaluation and cardiovascular response:An active coping approach. Journal of Personality and Social Psy-chology, 69, 530–543.

(Received March 26, 2008; Accepted July 28, 2008)

Incentives and preejection period 457