Emotions and Sport Performance:
An Exploration of Happiness,
Hope, and Anger
Tim Woodman, Paul A. Davis, Lew Hardy, Nichola Callow,
Ian Glasscock, and Jason Yuill-Proctor
We conducted three experiments to examine the relationships between emotions and
subcomponents of performance. Experiment 1 revealed that anger was associated
with enhanced gross muscular peak force performance but that happiness did not
inuence grammatical reasoning performance. Following Lazarus (1991, 2000a), we
examined hope rather than happiness in Experiment 2. As hypothesized, hope yielded
faster soccer-related reaction times in soccer players. Experiment 3 was an examina-
tion of extraversion as a moderator of the anger-performance relationship. When
angry, extraverts’ peak force increased more than introverts’. Results are discussed
and future research directions are offered in relation to Lazarus’s framework.
Keywords: happiness, hope, anger, performance, extraversion
Although a range of emotions has been observed in sport, including anxiety,
frustration, disappointment, happiness, hope, and anger (Crocker, Kowalski,
Graham, & Kowalski, 2002; Gould et al., 2000; Hanin, 2000; Jones, Lane, Bray,
Uphill, & Catlin, 2005; Jones & Uphill, 2004; Lazarus, 2000a; Robazza &
Bortoli, 2007; Sève, Ria, Poizat, Saury, & Durand, 2007), it is anxiety that has
received by far the most research attention. This is especially true for research
that has examined the emotion-performance relationship, which is the focus of
the present research.
Lazarus’s (1991, 2000a) cognitive-motivational-relational (CMR) theory
proposes that athletes’ specic emotions are each guided by a core relational
theme that describes the interaction between the individual and the environment.
The core relational theme is a summary of the appraisals that individuals make in
assessing the risk and reward involved in a particular situation. For example, the
core relational theme of anger is, “a demeaning offence against me and mine”
(Lazarus 2000a, p. 242). Each core relational theme has an associated action ten-
dency that directly represents the manifestation of the person’s appraisal of the
The authors are with the School of Sport, Health, and Exercise Sciences, Bangor University, Bangor,
U.K. A copy of all scripts is available from the rst author on request.
Journal of Sport & Exercise Psychology, 2009, 31, 169-188
© 2009 Human Kinetics, Inc.
170 Woodman et al.
stimulus in relation to the self (Lazarus, 2000b). The action tendency for anger is,
“a powerful impulse to counterattack in order to gain revenge for an affront or
repair a wounded self-esteem” (Lazarus, 2000a, p. 243).
Lazarus’s (1991, 2000a) CMR theory proposes that the core relational
theme and the associated action tendency will inuence performance depending
on the complex relationship between the athlete and the situation. For example,
anger may negatively impact performance if it draws resources away from the
primary task at hand. However, if the physical skill requires a “lashing out”
motion toward an aggressor or opponent, performance may be facilitated due to
its close association with anger’s action tendency (Lazarus, 2000b). As such,
Lazarus’s CMR theory offers a potentially fruitful theoretical framework for
investigating the likely complex emotion-performance relationship. Thus, it
seems surprising that it has received minimal performance research attention.
One reason for this may be that many emotions are best thought of as post-
performance emotions. For example, for positive emotions, a performer
happy to have won, relieved at having achieved a performance goal, proud to
have been in the nal of a major competition. Similarly, although anxiety can be
readily conceptualized as a pre-performance emotion, other negative emotions
are more readily thought of as postperformance emotions. For example, a per-
former may feel angry at having performed below potential, ashamed of a par-
ticularly poor performance in front of a crowd, guilty of letting the coach down
with a poor performance.
Although these emotions can be experienced postperformance (Gould et al.,
2000), some may also be experienced preperformance and may affect subse-
quent performance. In addition, there is evidence that positive affect is related
to a number of criterion variables, including health, marital well-being, rela-
tionship satisfaction, and coping (see Lyubomirsky, King, & Diener, 2005).
Further, in the sport domain, Totterdell (2000) found that happiness was posi-
tively related to cricket batting average. However, although Uphill and Jones
(2007) found some qualitative support for CMR theory, to the best of our knowl-
edge there are no studies that have studied the emotion-performance link within
Lazarus’s (2000a) CMR framework. The aim of the current study was to build
upon our understanding of the emotion-performance relationship within this
Physical and cognitive subcomponents of performance appear to be differen-
tially affected by emotional arousal (Partt, Hardy, & Pates, 1995; Partt, Jones, &
Hardy, 1990). For example, physiological arousal has been positively associated
with performance on aerobic tasks (Partt et al., 1995) and strength tasks (Perkins,
Wilson, & Kerr, 2001). In addition, heightened emotional intensity may sometimes
be benecial to performance especially if it can motivate individuals to invest
greater resources to the task at hand (cf. Eysenck & Calvo, 1992; Fredrickson,
2001; Lazarus, 2000a). Conversely, physiological arousal can be detrimental to
performance on tasks that require ne motor control (Noteboom, Fleshner, &
Enoka, 2001; Partt et al., 1990). Further, attempts to manage emotions have been
found to divert cognitive resources from the primary task toward coping strategies
(Janelle, 2002). Despite these promising research avenues, the majority of research
examining emotional arousal has focused on anxiety and has disregarded other
Emotions and Sport Performance 171
negative emotions (e.g., anger) as well as positive emotions (e.g., happiness, hope)
that display similar or different patterns of physiological activation (Jones, Lavallee,
& Thatcher, 2004; Lazarus, 2000b) that might differentially affect performance
(Robazza & Bortoli, 2007).
Given the previous promising research with happiness (e.g., Lyubomirsky
et al., 2005; Totterdell, 2000), we start with happiness as a positive emotion. In
addition, given the obvious theoretical benets of anger on a purely physical task
(i.e., the desire to lash out; Lazarus, 2000a), we explore anger as a negative emo-
tion. To this end, in Experiment 1 we explore the effect of anger and happiness on
the performance of physical and cognitive tasks. Specically, because the action
tendency for anger is associated with a lashing out movement (Lazarus, 2000a)
that is similar to the task requirements of a maximal force task, we hypothesize
that anger will benet performance on such a task. Conversely, happiness has
been shown to be positively related to effective problem solving (e.g., Erez &
Isen, 2002; Estrada, Isen, & Young, 1994; Kavanagh, 1987) and signals that all is
well and that resources can be committed to the task (Fredrickson, 2001; Lazarus,
2000b). Consequently, we hypothesize that happiness will facilitate performance
on the cognitive task.
Participants. Fifteen physically active students (9 men, 6 women; Mage = 24.18
years; SD = 3.75) participated in the experiment. All provided written informed
consent to participate in the experiment.
Imagery Scripts. Imagery scripts were composed for the purpose of inducing
happiness, anger, and an emotion-neutral affect. The emotion scripts (happiness
and anger) were based on Lazarus’s (1991, 2000a) core relational themes of hap-
piness and anger, and contained vivid detail regarding stimuli, response, and mean-
ing propositions to elicit physiological, cognitive, and somatic activation consistent
with the appropriate emotional state (Cumming, Olphin & Law, 2007; Lang, 1979).
The emotion-neutral script outlined the process of brushing one’s teeth (see
Kavanagh & Hausfeld, 1986). The delivery of the imagery scripts was standardized
by recording the scripts onto a compact disc.
Happiness and Anger. To assess the degree to which the emotions were experi-
enced, we presented participants with a Happiness and Anger inventory. Happiness
statements were derived from Gould et al.’s (2000) study, which examined athletes’
emotions during sport performance. The 10 happiness statements were chosen
using a deductive approach. These were I am ecstatic, I am happy, I feel elated, I
feel joyful, I feel blissful, I feel good, I feel pure happiness, I am on cloud nine, I am
full of joy, and I feel like smiling. The inventory also included the 10 state anger
statements (I am furious, I feel irritated, I feel angry, I feel like yelling at somebody,
172 Woodman et al.
I feel like breaking things, I feel mad, I feel like banging on the table, I feel like hit-
ting someone, I feel burned up, I feel like swearing) from the State-Trait Anger
Scale (STAS; Spielberger, Jacobs, Russell, & Crane, 1983). Each happiness and
anger statement was rated on a 4-point scale (1 = not at all, 4 = very much so).
Spielberger et al. (1983) reported high internal consistency with a Cronbach alpha
coefcient of .92. The Cronbach alphas for the current study were .86 for happiness
and .90 for anger.
Visual Analog Scale. Although we were exploring Lazarus’s (2000a) frame-
work, in which emotions are conceptualized as discrete, we also used a two-
dimensional Visual Analog Scale (VAS) to assess the degree to which participants
experienced the emotions of happiness and anger before undertaking the experi-
mental tasks. This was simply to verify further that the emotion manipulations had
been successful in inducing the appropriate emotions. Although such an approach
is in line with Russell’s (1980) circumplex model of affect, researchers, including
Lazarus (2000a) and Russell (2003), have questioned the usefulness of such two-
dimensional models. Indeed, based on this method alone, one would not be able
to differentiate between anxiety and anger, for example (Russell, 2003). However,
as an adjunct to the questionnaire data, we deemed this method appropriate for
gleaning additional discriminatory information about the success of our emotion
manipulations. We used a grid (two 200-mm axes each anchored by not at all and
very much) that measured orthogonally the dimensions of arousal and hedonic
Cognitive Task. We used a grammatical reasoning task that was originally devel-
oped by Baddeley (1968). This task requires participants to identify whether a
sentence describes a letter pair correctly (e.g., BA: A follows B; True or false?).
We presented participants with a list of 32 such pairs and asked them to complete
as many as possible in a 90-s period. Further, we told participants that they would
receive 1 point for every correct answer. Cognitive performance was assessed by
accuracy (i.e., number of correct responses). Because participants completed the
same task on three occasions, the questions were randomized into three different
orders (see Baddeley, 1968).
Physical Task. Participants performed a gross muscular peak force task on a Kin
Com Muscle Testing adjustable dynamometer (Model 125E+, Chattecx Corpora-
tion) as a measure of physical performance. Peak force (in newton meters) was
recorded by isometric extension of the right leg. After familiarization with the
equipment and task demands, participants kicked as fast and as hard as possible
for a period of 5 s. They performed the task twice with a period of 10 s between
the two trials. The mean of the two trials was used for analysis.
We informed participants that the experiment was an examination of performance
under different conditions of emotion and provided them with instructions on how
to complete the cognitive and physical tasks. We administered the experimental
conditions on different days at approximately the same time of day. Each partici-
pant completed the trials individually.
Emotions and Sport Performance 173
After providing demographic information and written informed consent, par-
ticipants sat at a desk and the experimenter outlined the emotion that was to be
induced during the testing session and asked participants to think of a situation in
which they had experienced this emotion. The corresponding imagery script was
then presented. When the imagery script had nished participants were asked to
indicate how they were feeling on the VAS. The cognitive task was then com-
pleted. Immediately after completing the task we asked participants to complete
the Happiness and Anger Inventory retrospectively in relation to how they had felt
during the task.
Participants then moved to the dynamometer to perform the physical task
under the same emotion condition. Upon conrmation that they were seated
securely, they performed one warm-up trial to familiarize themselves with the
task. We then presented the relevant imagery script. As soon as the imagery script
had nished, participants indicated how they felt on the VAS and were reminded
to perform the kick “as fast and as hard as you can.” Upon completion of the
physical task, the Happiness and Anger inventory was completed retrospectively.
To nish, we offered participants the inducement of a more pleasant emotion
(happiness) if they experienced residual unpleasant feelings (e.g., anger); no par-
ticipants required this service. We nished by thanking and debrieng the partici-
pants. The same procedure outlined above was followed for each condition (i.e.,
happiness, anger, and emotion-neutral). The order of presentation of the emotion
conditions was balanced and randomized across participants.
To assess the imagery scripts’ efcacy in inducing the respective emotions (i.e.,
happiness, anger, and neutral) during the physical and cognitive tasks, single-
factor repeated-measures ANOVAs were conducted to examine the dimensions of
arousal and pleasantness on the VAS, and the subscales of happiness and anger on
the Happiness and Anger inventory. One participant failed to provide data in the
emotion-neutral conditions and was removed from all analyses. When the assump-
tion of sphericity was violated, we applied a Greenhouse–Geisser adjustment to
the degrees of freedom.
Happiness and Anger. There was a signicant happiness difference across emo-
tion conditions, both for the cognitive task, F(1.17, 16.40) = 220.84, p < .001, 2 =
.94, and for the physical task, F(1.23, 17.25) = 313.06, p < .001, 2 = .96. Tukey’s
follow-up tests revealed that participants experienced signicantly more happi-
ness in the happiness condition compared with the anger and emotion-neutral
conditions (see Table 1). Moreover, there was a signicant anger difference across
emotion conditions for both the cognitive task, F(1.14, 16.01) = 54.88, p < .001,
2 = .80, and the physical task, F(1.12, 15.72) = 61.44, p < .001, 2 = .81. Tukey’s
follow-up tests revealed that participants experienced signicantly more anger in
the anger condition compared with the happiness and emotion-neutral conditions
(see Table 1).
Table 1 Visual Analog Scale (VAS) Arousal, VAS Pleasantness, Happiness, Anger, and Performance Means (SD)
for the Three Emotion Conditions in Experiment 1
Cognitive Task Physical Task
Happiness Anger Emotion-neutral Happiness Anger Emotion-neutral
Happiness ***40.33 (6.37) 11.33 (2.23) 29.00 (6.13) ***41.73 (5.32) 11.27 (2.37) 30.27 (5.15)
Anger 13.20 (7.03) ***38.60 (5.82) 25.53 (7.01) 12.93 (7.33) ***40.07 (5.62) 27.13 (6.83)
Arousal Intensity *7.15 (2.12) *6.96 (1.78) 3.79 (1.92) *7.11 (1.86) *6.80 (2.27) 3.64 (1.96)
Hedonic Tone ***8.92 (0.93) 2.18 (1.62) a5.39 (0.81) ***8.64 (1.05) 2.49 (1.93) a5.42 (1.10)
Performance 18.29 (5.53) 19.29 (4.78) 18.93 (5.68) 561.36 (190.66) *611.29 (203.85) 559.21 (191.75)
Note. Range of possible scores is as follows: arousal intensity, −10 to 10; hedonic tone, −10 to 10; happiness, 0 to 40; anger, 0 to 40.
*p < .05, *p < .01, ***p < .001.
a Signicantly greater than anger (p < .001).
Emotions and Sport Performance 175
Visual Analog Scale (VAS). The analyses revealed there was a signicant arousal
difference across emotion conditions, both for the cognitive task, F(2, 26) = 13.52,
p < .001, 2 = .51, and for the physical task, F(2, 26) = 12.48, p < .001, 2 = .49.
Tukey’s follow-up tests revealed that participants experienced signicantly greater
arousal in the happiness and anger conditions compared with the emotion-neutral
condition during both tasks (see Table 1); the anger and happiness conditions
were not signicantly different from each other during either of the tasks. Further,
there was a signicant hedonic tone (pleasantness) difference across emotions for
the cognitive task, F(2, 26) = 151.90, p < .001, 2 = .92, and for the physical task,
F(2, 26) = 73.53, p < .001, 2 = .85. Tukey’s follow-up tests revealed that partici-
pants experienced more pleasantness in the happiness condition compared with
the anger condition and less pleasantness in the anger condition compared with
the emotion-neutral condition during both tasks (see Table 1).
The combined results of the VAS and the Happiness and Anger inventory
reveal that the attempts to induce the respective emotions were successful. Further,
the VAS ndings suggest that the emotions of happiness and anger were character-
ized by high levels of arousal (Lazarus, 1991) and lend support to the proposal that
happiness is a more pleasurable emotion than anger (Russell, 1980).
Cognitive Task. A single-factor repeated-measures ANOVA revealed no signi-
cant difference across emotion conditions in the number of correct answers on the
grammatical reasoning task, F(1.43, 18.53) = .52, ns, 2 = .04 (see Table 1).
Physical Task. A single-factor repeated-measures ANOVA revealed a signicant
difference across emotion conditions for peak force, F(2, 26) = 4.52, p < .05, 2 =
.26. Tukey’s follow-up tests revealed that performance was signicantly greater in
the anger condition compared with the happiness and emotion-neutral conditions;
there was no signicant difference between the happiness and emotion-neutral
conditions (see Table 1).
The aim of Experiment 1 was to examine the inuence of anger and happiness on
cognitive and physical aspects of performance. The ndings partially supported
our hypotheses; participants’ performance on the physical task was signicantly
greater in the anger condition compared with the happiness and emotion-neutral
conditions. These results of the experiment are consistent with Lazarus’s (2000b)
suggestion that anger may facilitate physical performance if the required skill is
similar to anger’s associated action tendency (i.e., to lash out).
Although the anger results are encouraging, the results for happiness do not
support our hypothesis and previous research (e.g., Lyubomirsky et al., 2005;
Perkins et al., 2001; Totterdell, 2000). That is, happiness did not produce any sig-
nicant differences in cognitive performance. A possible explanation of the lack of
happiness ndings resides in the core relational theme for happiness: “making
reasonable progress toward the realization of a goal” (Lazarus, 2000a, p. 234).
This suggests that happiness may in fact result in no change in the cognitive
176 Woodman et al.
resources committed to the task. That is, the core relational theme of happiness
suggests a satiated state: happiness signals that all is well and there is possibly no
immediate need or desire to do anything to change this (see also Carver & Scheier,
1998; Mackie & Worth, 1989; Melton, 1995). Given these considerations, it is
perhaps not surprising that anger results in signicant performance gains (i.e., I
am angry; I want to lash out) and that happiness does not (i.e., I am happy; I do
not feel the need to do anything). We explore some alternative explanations of
these results in the general discussion following Experiment 3.
The aim of Experiment 2 was to investigate potential performance gains with a
more goal-oriented positive emotion. One obvious such emotion candidate is hope.
Indeed, the core relational theme for hope is “fearing the worst but yearning for
better, and believing the improvement is possible” (Lazarus, 2000a, p. 234; see also
Lazarus, 1999), which is more likely to result in greater mental effort. Further, hope
is a common preperformance emotion among athletes: despite fear of failure, they
hope for the best outcome. Given the core relational theme of hope, we hypothe-
sized that participants who were hopeful would believe that improved performance
was possible and would thus invest greater cognitive resources to the successful
completion of the task and perform better (Eysenck & Calvo, 1992; Lazarus, 2000a).
Conversely, as some cognitive resources might be diverted away from the primary
task toward coping strategies when participants were angry (Lazarus, 2000b), we
hypothesized that anger would not result in better performance on such a task.
There were two other potential limitations in Experiment 1. First, although
positive affect is thought to allow resources to be allocated to the task (cf. Fre-
drickson, 2001), we did not measure such resources. Second, the task (grammati-
cal reasoning) was of limited relevance to participants, which may have resulted
in few resources being allocated to the task. In Experiment 2 we aimed to redress
these limitations by developing a more sport-specic task for sport participants
and measuring resources via mental effort.
Participants. Eighteen semiprofessional male British soccer players (Mage =
21.50 years; SD = 2.12) participated in the experiment.
Imagery Scripts. Imagery scripts were used to elicit the emotional states of hope
and anger for the appropriate conditions (e.g., Cumming et al., 2007; Lang, 1979).
The imagery scripts were constructed in the same manner as in Experiment 1 but
with some specic reference to football, considering Lazarus’s (2000a) core rela-
tional themes for hope and anger. The emotion-neutral condition was the same as
in Experiment 1 (i.e., instructing participants to imagine brushing their teeth; cf.
Kavanagh & Hausfeld, 1986).
Hope and Anger. This inventory comprised nine hope statements (e.g., I am hop-
to do well on this task) derived from Gould et al. (2000) and the same 10 state
Emotions and Sport Performance 177
anger statements from the STAS (Spielberger et al., 1983) as were used in Experi-
ment 1. Each of the hope and anger statements was rated on a 5-point Likert-type
scale from 1 (not at all) to 5 (very much so). The hope statements were: I feel hope-
ful, I have hope, I am hoping to do well on this task, I feel hopeless about this task,
I have not got much hope, I have faith in my ability, I do not want to perform badly
on this task, I don’t really mind how I perform on this task, I hope I will perform
well on this task. The hope and anger subscales had high internal consistency with
Cronbach alpha coefcients of .89 and .88 for hope and anger, respectively.
The Sport Affect Grid. A Sport Affect Grid (SAG) assessed two independent dimen-
of affect: intensity and hedonic tone (pleasantness). The SAG has been used
previously in sport research (e.g., Hardy, Hall, & Alexander, 2001) and is pre-
sented as a 9 9 grid: the vertical axis assesses the self-perceived intensity of an
emotion, ranging from Extremely Low Intensity to Extremely High Intensity, and
the horizontal axis assesses hedonic tone, ranging from Unpleasant Feeling to
Pleasant Feeling. Participants are asked to mark an X on the part of the grid that
best represents how he/she feels right now. Scores for the intensity and hedonic
tone of the emotions were calculated separately by converting the location of the
X on each axis to a value from 1 to 9.
Cognitive Task. As the letter transformation task used in Experiment 1 has lim-
ited applicability to sport situations, we presented the soccer players with a task
that assessed their soccer-related reaction times. This computer task required par-
ticipants to track the path of an opposing player as closely as possible with the
cursor of the mouse while anticipating the appearance of a soccer ball on the
screen. When the soccer ball appeared on the screen, participants were to react as
quickly as possible by clicking the mouse. The task lasted 45 s with a total of nine
soccer balls appearing every 4, 5, or 6 s. The order of presentation of these time
periods was randomized within participants. Participants’ mean reaction time was
retained for analysis.
Perceived Mental Effort Scale (PMES; Mullen & Hardy, 2000). Given the core
relational theme of hope, we hypothesized that hope would be associated with an
increase in mental effort. Consequently, we asked participants to assess how much
mental effort they had invested in the task by completing the PMES: Based on the
most mental effort you have ever used to concentrate before, how would you rate
your concentration effort during your performance on the task? The PMES is
scored on a scale of 0 (No effort) through 5 (Moderately effortful) to 10 (Most
Participants attended the testing sessions individually and we told them that we
were studying emotions and soccer-related performance. After the participant
had provided written informed consent and demographic information, the
researcher explained the experimental task. Participants then sat at a desk in
front of a computer monitor and listened to the rst imagery script via head-
phones (i.e., hope, anger, or neutral). After the imagery script, participants com-
pleted the SAG and the computer soccer task. Immediately after the task,
participants retrospectively completed the Hope and Anger Inventory and the
178 Woodman et al.
PMES in relation to how they had felt immediately before the task. Once partici-
pants had completed the inventories and had rested for a few moments, we asked
participants to stand with their eyes closed and to balance on alternate legs while
counting backward in threes from 100 to zero. This was performed between
each of the conditions to minimize any carryover effects from one emotion con-
dition to the next.
The order of the conditions (i.e., hope, anger, and neutral) was balanced
across participants. After the third condition, we thanked participants for their
time, offered them the opportunity to ask any questions, and debriefed them before
they left the laboratory.
Hope and Anger. Single-factor repeated-measures ANOVAs were conducted across
the three emotion conditions (i.e., hope, anger, neutral). There was a signicant
hope difference across conditions, F(2, 34) = 4.60, p < .05, 2 = .21. Tukey’s
follow-up tests revealed that signicantly more hope was expressed in the hope
condition compared with the anger and emotion-neutral conditions; there was no
signicant difference in hope between the anger and emotion-neutral conditions.
Further, there was a signicant anger difference across conditions, F(1.40, 23.85)
= 31.50, p < .001, 2 = .65. Tukey’s follow-up tests revealed that signicantly
more anger was reported in the anger condition compared with the hope and emo-
tion-neutral conditions; there was no signicant difference in anger between the
hope and emotion-neutral conditions (see Table 2).
The Sport Affect Grid. Single-factor repeated-measures ANOVAs were conducted
to examine the dimensions of hedonic tone (pleasantness) and intensity across
emotions. There was a signicant hedonic tone difference, F(2, 34) = 11.37,
Table 2 Hope, Anger, Hedonic Tone, Intensity, Mental Effort,
and Mean Reaction Times (SD) for the Three Emotion Conditions
in Experiment 2
Hope Anger Emotion-neutral
Hope 40.11* (3.83) 35.33 (7.97) 32.67 (7.77)
Anger 16.33 (6.58) 31.11* (11.72) 11.94 (2.55)
Hedonic Tone 6.94* (2.10) 4.06 (2.28) 5.94* (1.43)
Arousal 7.56* (1.24) 7.94* (1.58) 3.61 (2.00)
Mental Effort 8.06* (1.39) 7.78* (1.48) 4.78 (1.77)
Reaction Time 420.70* (104.73) 424.29 (75.34) 448.91 (88.83)
Note. Range of possible scores is as follows: anger, 0 to 50; hope, 0 to 45; hedonic tone, 1 to 9; arousal,
1 to 9; mental effort, 0 to 10.
* p < .05.
Emotions and Sport Performance 179
p < .001, 2 = .40. Tukey’s follow-up tests revealed that the anger condition
yielded signicantly lower pleasantness than the hope and neutral conditions. The
difference between the hope and emotion-neutral conditions approached conven-
tional signicance with greater hedonic tone expressed in the hope condition, p =
.09. Further, there was a signicant intensity difference across emotions, F(2, 34)
= 40.71, p < .001, 2 = .71. Tukey’s follow-up tests revealed that both the anger
and hope conditions were signicantly more intense than the emotion-neutral
condition with no signicant difference between the anger and hope conditions
(see Table 2).
These analyses suggest the imagery scripts were successful in inducing the
corresponding emotions. Further, they conrm that anger is an unpleasant emo-
tion that is characterized by a high level of intensity (Lazarus, 2000a; Russell,
1980) and that hope is a pleasant and intense emotion (Lazarus, 2000a).
Mental Effort. A single-factor repeated-measures ANOVA revealed that partici-
pants’ mental effort differed across conditions, F(2, 34) = 30.75, p < .001, 2 =
.64. Tukey’s follow-up tests showed that the mental effort invested in the hope and
anger conditions was signicantly greater than in the emotion-neutral condition;
there was no signicant difference between the mental effort invested in the hope
and anger conditions (see Table 2).
Reaction Time. A one-way repeated-measures ANOVA revealed no signicant
difference between emotion conditions on reaction time, F(2, 34) = 2.12, p = .14,
2 = .11. However, given that our hypothesis was that hope would yield faster
reaction times than no emotion and that anger and emotion-neutral conditions
would not differ, we proceeded with these two a priori comparisons. These
revealed that the reaction times in the hope condition were signicantly faster than
in the emotion-neutral condition, t(17) = 2.47, p < .05, 2 = .26, and that the anger
and emotion-neutral conditions did not signicantly differ, t(17) = 1.69, p = .11,
2 = .14 (see Table 2).
The purpose of Experiment 2 was to examine the inuence of hope and anger on
cognitive performance. The ndings largely supported our hypothesis. That is,
effort and performance were greater in the hope condition compared with the
emotion-neutral condition. In the anger condition, although there was a signi-
cant increase in effort, performance was not signicantly improved compared
with the emotion-neutral condition. This is possibly because the core relational
theme of hope (i.e., yearning for better) can be directed to the task at hand. In this
way, the hope-associated increase in mental effort was accompanied by an
increase in performance, thereby rendering hope the more efcient emotion on
this largely cognitive task. Indeed, the action tendency for anger (e.g., lashing
out) can less obviously be directed to the cognitive task (cf. Lazarus, 2000a),
which is possibly why the anger-associated increase in mental effort was not
translated into a signicant increase in performance. For anger to be an effective
180 Woodman et al.
performance-enhancing emotion it appears that the task needs to be closely
aligned with anger’s action tendency (e.g., lashing out; Lazarus, 2000a). This was
demonstrated in Experiment 1, in which anger was associated with better perfor-
mance on a maximal force gross muscular task. We discuss the similarity of the
anger and hope reaction times further in the general discussion.
The aim of Experiment 3 was to further our understanding of how individ-
ual differences might moderate the emotion-performance relationship and spe-
cically this anger-performance relationship. It has long been established (e.g.,
Hanin, 1980, 2000) that individuals’ performance will be affected by their emo-
tional state. For example, Hanin’s individual zones of optimal functioning
(IZOF) model states that individuals will perform better when they are within
their preferred emotional range. In its simplest form, the model predicts that
people are different and that their emotions will affect performance differently.
Although this is helpful in an applied context, it bears limited theoretical weight.
As Gould and Tuffey (1996) noted, the IZOF model is an individual difference
model without any individual difference variables. In Experiment 3, we sought
to explore a more theoretically derived individual difference approach to the
With specic reference to anger in the context of developing the results of
Experiment 1, extraversion as an individual difference variable appears an obvious
potential moderator candidate. Extraverts are sociable and active
people who will more willingly express themselves in front of others (Goldberg,
1992, 1993), and recent research has revealed a facilitative extraversion and emo-
tional expression effect on performance in the cognitive domain (Perbandt, 2007).
Further, Cerin (2004) found that individuals higher in extraversion interpreted
their anxiety as more facilitative than individuals lower in extraversion. However,
there has been limited research examining the role of emotional expression and
personality on physical performance. As Experiment 1 conrmed a facilitative
performance effect for anger, it follows that extraverts’ willingness to express
their anger should translate into greater performance benets. In other words, the
performance-related benets of expressing anger will be greater for extraverts.
This is the hypothesis of Experiment 3.
Participants. Seventy-two physically active undergraduate students (45 men, 27
women; Mage = 22.23 years; SD = 3.68) participated in the experiment.
Imagery Scripts. The imagery scripts from Experiment 1 were used to elicit the
appropriate emotions for the anger and emotion-neutral conditions (i.e., brushing
State-Trait Anger Scale. Participants completed the state section of the STAS
(Spielberger et al., 1983). In each condition, participants were asked to complete
Emotions and Sport Performance 181
the scale in relation to how they had felt after hearing the emotional induction
script immediately before the task. Each of the anger statements was rated on a
4-point scale (1 = not at all, 4 = very much so). The Cronbach alpha coefcient for
the state anger subscale was .92.
Visual Analog Scale. The VAS from Experiment 1 was used to measure the
degree to which participants felt the dimensions of arousal and hedonic tone
(pleasantness) before performing the task.
Extraversion. Participants completed the International Personality Item Pool
(IPIP; Goldberg, 1993). The IPIP assesses individuals’ Big-Five personality mark-
ers: Extraversion, Agreeableness, Conscientiousness, Emotional Stability, and
Intellect/Imagination. The present experiment used the 50-item version consisting
of 10 items for each of the Big-Five personality factors. Participants were asked
to read each statement and then to rate how well it described them on a 5-point
scale from 1 (very inaccurate) to 5 (very accurate). We used only the data from
the extraversion scale (e.g., I am the life of the party) for analysis. The Cronbach
alpha coefcient for the extraversion subscale was .91.
Physical Performance. Participants performed individually the gross muscular
peak force task used in Experiment 1.
We told participants that the experiment was an examination of emotion and per-
formance and provided instructions on how to complete the task. After providing
demographic information and written informed consent, participants completed
the IPIP. We then secured the participant to the dynamometer and he/she com-
pleted the strength task following the same procedure as in Experiment 1. The
emotions were induced as in Experiments 1 and 2. Participants completed the task
individually under anger and emotion-neutral conditions on different days at
approximately the same time of day. The order of presentation of the emotion
conditions was balanced across participants.
State-Trait Anger Scale. A paired samples t test on the STAS revealed that anger
was signicantly greater in the anger condition (M = 27.57, SD = 6.95) than in the
emotion-neutral condition (M = 11.54, SD = 3.74), t(71) = 19.11, p < .001, 2 = .84.
Visual Analog Scale. We conducted paired samples t tests to examine the dimen-
sions of arousal and pleasantness across the two emotion conditions (anger, neu-
tral). Arousal in the anger condition (M = 144.57, SD = 29.45) was signicantly
greater than in the emotion-neutral condition (M = 77.94, SD = 43.57), t(71) =
11.10, p < .001, 2 = .63. Pleasantness in the anger condition (M = 75.83, SD =
40.86) was signicantly lower than in the emotion-neutral condition (M = 114.96,
SD = 40.31), t(71)= 6.43, p < .001, 2 = .37.
182 Woodman et al.
A paired samples t test revealed that the mean performance score in the anger
condition (377.97, SD = 135.86) was signicantly greater than in the emotion-
neutral condition (301.94, SD = 111.16), t(71) = 8.99, p < .001, 2 = .53. This
replicates the ndings of Experiment 1. To examine the extent to which extraver-
sion allows individuals to glean additional anger-induced performance benets,
we created a performance improvement score (i.e., the ratio of anger and neutral
performance scores). Extraversion was signicantly related to this improvement
score, r = .21, p < .05, thus suggesting that extraversion facilitates anger-induced
To further investigate the potential moderating role of extraversion in the
anger-performance relationship, we conducted a 2 (emotion: anger, neutral) 2
(extraversion: high, low) mixed-model ANOVA with repeated measures on the
rst factor and with a median split on the extraversion data. This conrmed a sig-
nicant main effect for emotion condition, F (1, 70) = 82.73, p < .001, 2 = .54,
such that participants performed better in the anger condition than in the emotion-
neutral condition. Of more central interest, the ANOVA revealed a signicant
interaction between emotion and extraversion, F (1, 70) = 6.90, p < .05, 2 = .09,
which conrmed that the performance benets when angry were signicantly
greater for extraverts than for introverts. One could argue that performing a median
split on the extraversion data are insufcient for classifying individuals as extra-
verts or introverts. Consequently, we conducted the same analysis using quartile
splits on the extraversion data. This yielded the same pattern of results. Speci-
cally, a signicant main effect for emotion condition, F (1, 36) = 50.69, p < .001,
2 = .59, and a signicant extraversion x emotion condition interaction, F (1, 36)
= 4.71, p < .05, 2 = .12. This interaction is illustrated in Figure 1.
The anger ndings replicated those of Experiment 1. That is, anger resulted in
signicantly greater performance on a gross muscular task. Further, in support of
our hypothesis, extraverts’ performance gains were greater than introverts’. These
ndings are also consistent with recent research examining the inuence of per-
sonality and emotional expression on cognitive performance and behavior (e.g.,
Perbandt, 2007; Smits & De Boeck, 2007; Smulders & Meijer, 2008).
Although the role of emotion in sport performance has been widely recognized
(Hanin, 2000; Lazarus, 2000a), only limited research has examined the perfor-
mance effects of emotions beyond those of anxiety. The aim of the present research
was to extend our understanding of the emotion-performance relationship by
investigating the effect of specic emotions on physical and cognitive aspects of
The ndings largely support Lazarus’s (2000a) theoretical framework. Spe-
cically, if the emotion experienced is aligned with the task demands then it seems
Emotions and Sport Performance 183
to facilitate performance. The anger ndings are also consistent with applied
research on anger in combative and contact sports (e.g., Robazza & Bortoli, 2007;
Terry & Slade, 1995). Although hope increased mental effort and reduced reaction
time, happiness did not improve performance on a different cognitive task, pos-
sibly because it reects a satiated state where no increase in effort is deemed
necessary (also possibly because of task differences, which we address later). As
such, in relation to performance, happiness may be most relevant as a postperfor-
The present line of research is clearly in its infancy. Indeed, there is a wealth
of research that anxiety researchers have promulgated: the multidimensional
nature of anxiety (e.g., Martens, Vealey, & Burton, 1990); different theoretical
frameworks including processing efciency (Eysenck & Calvo, 1992), conscious
processing (Masters, 1992), and catastrophe models (Hardy, 1990); the facilitative
and debilitative nature of anxiety (Jones, Hanton, & Swain, 1994); the frequency
of anxiety-related cognitive intrusions (Hanton, Thomas, & Maynard, 2004); as
well as the complex interplay between stress, anxiety, and performance (e.g.,
Woodman & Hardy, 2001). However, despite promising research ndings, there is
a relative paucity of theoretically driven research available on other emotions as
they relate to sport performance. Furthermore, with specic reference to the cur-
rent study, we have only investigated linear effects between emotion and perfor-
mance. Nonlinear relationships is an area that future researchers would do well to
address, as there are likely emotion thresholds beyond which the emotion no
longer facilitates performance and likely debilitates it, perhaps in a catastrophic
manner similar to anxiety (see Hardy, 1990).
Also, as evidenced in Experiment 2, there are likely facets to positive emo-
tions (e.g., hope) that are different to those of negative emotions (e.g., anger)
Figure 1 — The signicant interaction between emotion condition and extraversion on the
maximal force task in Experiment 3.
184 Woodman et al.
that may allow the performer to persist longer at a task or to persist more ef-
ciently, for example. Specically, mental effort played a signicant role in
Experiment 2, in which participants reported the greatest investment of mental
effort in the hope condition (albeit not signicantly different from effort in the
anger condition). Of course, any associated performance differences across
emotions are not likely to be simply categorized by negative and positive emo-
tions, as these are likely intertwined (cf. Levine, 1996). For example, an athlete
may hope to win a gold medal while simultaneously fearing that she might have
a disaster (cf. Lazarus, 2000a).
Although only hope yielded signicantly better performance in comparison
with the emotion-neutral condition, the performance difference between the hope
and anger emotion conditions was fairly minimal (see Table 2). In addition, both
hope and anger yielded an increase in effort invested on the task. Thus, it appears
that hope and anger may not be that different in their effects on reaction time. This
begs the question: would any arousal-inducing emotion have resulted in an
increase in effort and, if so, would any associated effort always be benecial to
performance? This seems unlikely. For example, a conscious processing view
(e.g., Masters, 1992) would suggest that effort will debilitate performance if such
effort is self-directed whereas other control process views (e.g., Eysenck & Calvo,
1992) suggest that effort can serve a regulatory function and help performance.
Although these theories have anxiety as their basis, other emotions may be worthy
of investigation within similar frameworks.
Extraversion signicantly moderated the degree to which participants experi-
enced anger-derived performance benets. This is promising because there is a
paucity of performance-related research that incorporates athletes’ personality. In
the current study, we operationalized anger as an intense and unpleasant emotion
(Lazarus, 1991; Russell, 1980) and measured it using Spielberger et al.’s (1983)
State Anger Inventory. However, there is increasing evidence that anger is a rather
more complex emotion (see Russell & Fehr, 1994) including two distinct anger
coping styles, most often referred to as anger-in and anger-out (Averill 1983;
Smits & Kuppens, 2005; Spielberger, Reheiser, & Sydeman, 1995). The anger-in
style refers to the person who experiences anger but keeps the expression of this
anger in. The anger-out style refers to the person who experiences anger and
expresses it outward. However, in the specic context of the present data, these
anger styles may reect the more fundamental personality construct of extraver-
sion. The degree to which anger expression style (i.e., anger-in and anger-out)
may mediate or moderate the extraversion interaction revealed here seems a
worthy avenue for future research on anger-performance and the role of individual
differences in that relationship.
Limitations and Future Research
Although the cognitive task used in Experiment 2 had greater ecological validity
than that used in Experiment 1, neither have strong ecological validity. Future
research employing more ecologically valid tasks might allow us to better
stand how emotion affects sport-specic performance. Such research
might include testing pre-event naturally occurring emotions and their effects on
Emotions and Sport Performance 185
subsequent subcomponents of performance. This approach would parallel the
approach used by Hardy and associates in earlier anxiety research (see, for
example, Partt et al., 1990).
On the basis of the null ndings for happiness in Experiment 1, we sug-
gested that happiness might be better investigated as a postperformance emotion.
Indeed, although the happiness manipulation was successful in inducing intense
and positive affect, participants’ performance did not benet from such states,
which was contrary to our hypotheses and somewhat contrary to previous studies
(e.g., Lyubomirsky et al., 2005; Perkins et al., 2001). We initially attempted to
explain this nding as happiness reecting a satisfactory status quo; that is, the
happy person feels little need to actively change anything (cf. Mackie & Worth,
1989; Melton, 1995). On this basis, we abandoned happiness in favor of hope in
Experiment 2. However, a closer look at Experiment 1 suggests that this aban-
donment may have been rather premature; another possible explanation for the
null ndings is that the task used was such that the motivation to engage in it was
insufcient. This is quite possible for two reasons. First, the reasoning task was
likely not of great interest to participants and certainly not of any direct rele-
vance. Past research (e.g., Lyubomirsky et al., 2005) has found that positive
affect does lead to success across a plethora of life domains (e.g., marriage,
health), which hold considerably more personal meaning to people than a gram-
matical reasoning task, for example. In response to this lack of task relevancy, in
Experiment 2 we attempted to align the task demands (soccer-related task) with
the sample (semiprofessional soccer players) and the effect of hope on perfor-
mance was evident. Future research would do well to investigate similar para-
digms with happiness to ascertain the degree to which happiness might affect
performance on tasks that hold more personal meaning for the individual. Second,
the task may have been insufciently challenging to motivate participants. For
example, Erez and Isen (2002) found that happiness was associated with better
performance but that happiness was motivational only when the task had reached
a certain degree of difculty. This suggests that happiness may only affect per-
formance when it matters most: “when the going gets tough.” That is, positive
affect may provide both sufcient resources and sufcient motivation to pursue
a demanding task (Fredrickson, 2001). This seems particularly worthy of future
research in the context of sport performance.
A nal limitation in relation to the happiness experiment is sample size,
which again suggests that happiness warrants further research attention. Of course,
sample size does not explain the hypothesized ndings that were obtained for the
other emotions (i.e., hope and anger), which suggests that this was not a major
shortcoming of these experiments. Further, the anger results were replicated
across the experiments despite potential sample size concerns in Experiment 1.
The experiments reported here offer support for the notion that emotions
other than anxiety are worthy of research attention when attempting to under-
stand the emotion-performance relationship. However, there is an important
shortcoming here (and in all other research on emotions and performance) that
is worth considering in future. That is, we did not control anxiety. In other
words, we cannot be certain that the anger-performance relationship revealed
here does not simply reect an anxiety nding and that all negative, arousal-
inducing, emotions would yield similar results. Future research would do well
186 Woodman et al.
to investigate anxiety and anger in conjunction with each other to ascertain the
degree to which these emotions might yield differential, additive, or interactive
effects on performance.
In summary, Lazarus’s (2000a) theoretical framework appears to be a promis-
ing avenue for researchers interested in the effect of emotions on performance.
Happiness did not affect performance, which may reect a self-satised state
although future research needs to explore more demanding and relevant task per-
formance. Hope facilitates performance on reaction time tasks and anger helps
performance of gross muscular tasks. Furthermore, extraverts benet most from
such anger-induced physical performance increases. The results of the present
studies suggest that emotions other than anxiety deserve further attention.
The authors would like to thank two anonymous reviewers for their helpful and insightful
comments and suggestions on an earlier draft of this manuscript.
Averill, J.R. (1983). Studies on anger and aggression: Implications for theories of emotion.
The American Psychologist, 38, 1145–1160.
Baddeley, A. (1968). A three minute reasoning test based on grammatical transformation.
Psychonomic Science, 10, 341–342.
Carver, C.S., & Scheier, M.F. (1998). On the self-regulation of behavior. New York: Cam-
bridge University Press.
Cerin, E. (2004). Predictors of competitive anxiety direction in male Tae Kwon Do practi-
tioners: A multilevel mixed idiographic/nomothetic interactional approach. Psychol-
ogy of Sport and Exercise, 5, 497–516.
Crocker, P.R.E., Kowalski, K.C., Graham, T.R., & Kowalski, N.P. (2002). Emotion in sport.
In J.M. Silva, III & D.E. Stevens (Eds.), Psychological foundations of sport (pp. 107–
131). Boston, MA: Allyn & Bacon.
Cumming, J., Olphin, T., & Law, M. (2007). Self-reported psychological states and physi-
ological responses to different types of motivational general imagery. Journal of Sport
& Exercise Psychology, 29, 629–644.
Erez, A., & Isen, A.M. (2002). The inuence of positive affect on the components of expec-
tancy motivation. The Journal of Applied Psychology, 87, 1055–1067.
Estrada, C.A., Isen, A.M., & Young, M.J. (1994). Positive affect inuences creative prob-
lem solbving and reported source of practice satisfaction in physicians. Motivation
and Emotion, 18, 285–299.
Eysenck, M.W., & Calvo, M.G. (1992). Anxiety and performance: The processing ef-
ciency theory. Cognition and Emotion, 6, 409–434.
Fredrickson, B.L. (2001). The role of positive emotions in positive psychology: The broaden-
and-build theory of positive emotions. The American Psychologist, 56, 218–226.
Goldberg, L.R. (1992). The development of markers for the Big-Five factor structure. Psy-
chological Assessment, 4, 26–42.
Goldberg, L.R. (1993). The structure of phenotypic personality traits. The American Psy-
chologist, 48, 26–34.
Gould, D., Medberry, R., Dieffenbach, K., Lauer, L., Hardy, L., & Jones, G. (2000, Octo-
ber). Identifying the Range of Emotions Involved in Sport Achievement Striving. Poster
presented at the annual meeting of the Association for the Advancement of Applied
Sport Psychology, Vancouver, Canada.
Emotions and Sport Performance 187
Gould, D., & Tuffey, S. (1996). Zones of optimal functioning research: A review and cri-
tique. Anxiety, Stress, &. Coping, 9, 53–68.
Hanin, Y.L. (1980). A cognitive model of anxiety in sports. In W. Straub (Ed.), Sport psy-
chology: An analysis of athlete behavior (pp. 236–249). Ithaca, NY: Movement.
Hanin, Y.L. (2000). Successful and poor performance and emotions. In Y.L. Hanin (Ed.),
Emotions in sport (pp. 157–187). Champaign, IL: Human Kinetics.
Hanton, S., Thomas, O., & Maynard, I. (2004). Competitive anxiety responses in the week
leading up to competition: The role of intensity, direction and frequency dimensions.
Psychology of Sport and Exercise, 5, 169–181.
Hardy, J., Hall, C., & Alexander, M. (2001). Exploring self-talk and affective states in
sport. Journal of Sports Sciences, 19, 469–475.
Hardy, L. (1990). A catastrophe model of anxiety and performance. In J.G. Jones & L.
Hardy (Eds.), Stress and performance in sport (pp. 81–106). Chichester, UK: Wiley.
Janelle, C.M. (2002). Anxiety, arousal and visual attention: A mechanistic account of per-
formance variability. Journal of Sports Sciences, 20, 237–251.
Jones, J.G., Hanton, S., & Swain, A.B.J. (1994). Intensity and interpretation of anxiety
symptoms in elite and non-elite sports performers. Personality and Individual Differ-
ences, 17, 657–663.
Jones, M.V., Lane, A.M., Bray, S.R., Uphill, M., & Catlin, J. (2005). Development and
validation of the sport emotion questionnaire. Journal of Sport & Exercise Psychol-
ogy, 27, 407–431.
Jones, M.V., Lavallee, D., & Thatcher, J. (2004). Coping and emotion in sport: Future
directions. In D. Lavalee, J. Thatcher, & M.V. Jones (Eds.), Coping and Emotion in
Sport (pp. 273–280). New York, NY: Nova Science Publishers Inc.
Jones, M.V., & Uphill, M.A. (2004). Emotion in sport: Antecedents and performance con-
sequences. In D. Lavalee, J. Thatcher, & M.V. Jones (Eds.), Coping and Emotion in
Sport (pp. 9–28). New York: Nova Science Publishers Inc.
Kavanagh, D., & Hausfeld, S. (1986). Physical performance and self-efcacy under happy
and sad moods. Journal of Sport Psychology, 8, 112–123.
Kavanagh, D.J. (1987). Mood, persistence, and success. Australian Journal of Psychology,
Lang, P.J. (1979). A bio-informational theory of emotional imagery. Psychophysiology,
Lazarus, R.S. (1991). Cognition and motivation in emotion. The American Psychologist,
Lazarus, R.S. (1999). Hope: An emotion and a vital coping resource against despair. Social
Research, 66, 653–678.
Lazarus, R.S. (2000a). How emotions inuence performance in competitive sports. The
Sport Psychologist, 14, 229–252.
Lazarus, R.S. (2000b). Cognitive-motivational-relational theory of emotion. In Y.L. Hanin
(Ed.), Emotions in sport (pp. 39–63). Champaign, IL: Human Kinetics.
Levine, L.J. (1996). Anatomy of disappointment: A naturalistic test of appraisal models of
sadness, anger, and hope. Cognition and Emotion, 10, 337–359.
Lyubomirsky, S., King, L., & Diener, E. (2005). The benets of frequent positive affect:
Does happiness lead to success? Psychological Bulletin, 131, 803–855.
Mackie, D.M., & Worth, L.T. (1989). Processing decits and the mediation of positive
affect in persuasion. Journal of Personality and Social Psychology, 57, 27–40.
Martens, R., Vealey, R.S., & Burton, D. (1990). Competitive anxiety in sport. Champaign,
IL: Human Kinetics.
Masters, R.S.W. (1992). Knowledge, knerves, and know-how. The British Journal of Psy-
chology, 83, 343–358.
Melton, R.J. (1995). The role of positive affect in syllogism performance. Personality and
Social Psychology Bulletin, 21, 788–794.
188 Woodman et al.
Mullen, R., & Hardy, L. (2000). State anxiety and motor performance: Testing the con-
scious processing hypothesis. Journal of Sports Sciences, 18, 785–799.
Noteboom, J.T., Fleshner, M., & Enoka, R.M. (2001). Activation of the arousal response
can impair performance on a simple motor task. Journal of Applied Physiology, 91,
Partt, G., Hardy, L., & Pates, J. (1995). Somatic anxiety and physiological arousal: Their
effects upon a high anaerobic, low memory demand task. International Journal of
Sport Psychology, 26, 196–213.
Partt, C.G., Jones, J.G., & Hardy, L. (1990). Multidimensional anxiety and performance.
In J.G. Jones & L. Hardy (Eds.), Stress and performance in sport (pp. 43–80). Chich-
ester, England: Wiley.
Perbandt, K. (2007). Differential effects of incongruent anger expression on memory per-
formance. Journal of Individual Differences, 28, 129–137.
Perkins, D., Wilson, G.V., & Kerr, J.H. (2001). The effects of elevated arousal and mood
on maximal strength performance in athletes. Journal of Applied Sport Psychology,
Robazza, C., & Bortoli, L. (2007). Perceived impact of anger and anxiety on sporting per-
formance in rugby players. Psychology of Sport and Exercise, 8, 875–896.
Russell, J.A. (1980). A circumplex model of affect. Journal of Personality and Social Psy-
chology, 39, 1161–1178.
Russell, J.A. (2003). Core affect and the psychological construction of emotion. Psycho-
logical Review, 110, 145–172.
Russell, J.A., & Fehr, B. (1994). Fuzzy concepts in a fuzzy hierarchy: Varieties of anger.
Journal of Personality and Social Psychology, 67, 186–205.
Sève, C., Ria, L., Poizat, G., Saury, J., & Durand, M. (2007). Performance-induced emo-
tions experienced during high-stakes table tennis matches. Psychology of Sport and
Exercise, 8, 25–46.
Smits, D.J., & De Boeck, P. (2007). From anger to verbal aggression: Inhibition at different
levels. Personality and Individual Differences, 43, 47–57.
Smits, D.J., & Kuppens, P. (2005). The relations between anger, coping with anger, and
aggression, and the BIS/BAS system. Personality and Individual Differences, 39,
Smulders, F., & Meijer, E. (2008). Extraversion and performance: A cognitive-energetical
approach. Personality and Individual Differences, 44, 475–486.
Spielberger, C.D., Jacobs, G., Russell, J.S., & Crane, R.S. (1983). Assessment of anger:
The State-Trait Anger Scale. In J.N. Butcher & C.D. Spielberger (Eds.), Advances in
Personality Assessment (pp. 161–187). London: Wiley.
Spielberger, C.D., Reheiser, E.C., & Sydeman, S.J. (1995). Measuring the experience,
expression, and control of anger. In H. Kassinove (Ed.), Anger disorders: Denitions,
diagnosis, and treatment (pp. 49–67). Washington, DC: Taylor & Francis.
Terry, P., & Slade, A. (1995). Discriminant effectiveness of psychological state measures in
predicting performance outcome in karate competition. Perceptual and Motor Skills,
Totterdell, P. (2000). Catching moods and hitting runs: Mood linkage and subjective perfor-
mance in professional sports teams. The Journal of Applied Psychology, 83, 848–859.
Uphill, M.A., & Jones, M.V. (2007). Antecedents of emotions in elite athletes: A cogni-
tive motivational relational theory perspective. Research Quarterly for Exercise and
Sport, 78, 79–89.
Woodman, T., & Hardy, L. (2001). Stress and anxiety. In R.N. Singer, H.A. Hausenblas,
& C.M. Janelle (Eds.), The Handbook of Sport Psychology (pp. 290–318). New York:
Manuscript received: June 4, 2008
Revision accepted: October 29, 2008