ArticlePDF Available

Abstract

Objectives In the context of Grand Slam tennis, we sought to examine how situational pressure and prior errors can disrupt subsequent performance in elite performers. Methods A retrospective analysis of more than 650,000 points across 12 Grand Slam tennis tournaments from 2016 to 2019 was conducted to identify pressurised in-game moments and unforced errors. A scoring system was used to index situational pressure based on the current match situation (e.g., break points, stage of the match) on a point-by-point basis. The occurrence of performance errors was identified based on double faults and unforced errors, as instances of controllable mistakes. Results A mixed effects logistic regression model revealed that an increase in the pressure index (a 1–5 score) significantly increased the probability of a performance error (ps < .001), as did an error on the preceding point (OR = 1.2, 95%CI [1.17, 1.23], p < .001). A multiplicative effect of pressure and prior errors also emerged, as the negative impact of prior errors on performance was greater when situational pressure was already high, in line with the predictions of Attentional Control Theory: Sport (ACTS). Analyses of the distribution of winners and unforced errors across individual players revealed that winning players were as susceptible to pressure and prior errors as losing players. Conclusions These findings extend our understanding of how ongoing feedback from prior mistakes may further exacerbate the effects of pressure on performance.
Accepted for publication in Psychology of Sport and Exercise 25/05/21 PREPRINT
1
Psychological pressure and compounded errors during elite-level tennis
David J. Harris
University of Exeter
Michael W. Eysenck,
Royal Holloway University of London
Samuel J. Vine,
University of Exeter
Mark R. Wilson
University of Exeter
Objectives: In the context of Grand Slam tennis, we sought to examine how situational
pressure and prior errors can disrupt subsequent performance in elite performers.
Methods: A retrospective analysis of more than 650,000 points across 12 Grand Slam
tennis tournaments from 2016-2019 was conducted to identify pressurised in-game
moments and unforced errors. A scoring system was used to index situational pressure
based on the current match situation (e.g., break points, stage of the match) on a point-
by-point basis. The occurrence of performance errors was identified based on double
faults and unforced errors, as instances of controllable mistakes. Results: A mixed
effects logistic regression model revealed that an increase in the pressure index (a 1-5
score) significantly increased the probability of a performance error (ps<.001), as did an
error on the preceding point (OR=1.2, 95%CI [1.17, 1.23], p<.001). A multiplicative
effect of pressure and prior errors also emerged, as the negative impact of prior errors on
performance was greater when situational pressure was already high, in line with the
predictions of Attentional Control Theory: Sport (ACTS). Analyses of the distribution of
winners and unforced errors across individual players revealed that winning players were
as susceptible to pressure and prior errors as losing players. Conclusions: These findings
extend our understanding of how ongoing feedback from prior mistakes may further
exacerbate the effects of pressure on performance.
Keywords; anxiety; dependency; choking; failure; clutch;
The competitive sporting environment generates psychological pressure, described as ‘any
factor or combination of factors that increases the importance of performing well’ (Baumeister, 1984,
pp. 610). These factors include performance-contingent rewards, competition, ego relevance, and
audience observation (Baumeister & Showers, 1986). Research in sport psychology has documented
Accepted for publication in Psychology of Sport and Exercise 25/05/21 PREPRINT
2
how psychological pressure can create the negative emotional state of anxiety (comprised of cognitive
worry and physiological arousal; Eysenck, 2013), which subsequently impairs the execution of well-
learned skills (Roberts et al., 2019). This impairment has been termed choking and occurs when a
performer exhibits a negative response to perceived pressure, despite striving to perform well
(Baumeister, 1984; Beilock & Gray, 2007; Hill et al., 2010). Irrespective of whether this breakdown
occurs due to the disruption of automated motor processes or distraction via worry (see, (Payne et al.,
2018; Roberts et al., 2017)for recent reviews), there remains a persistent puzzle surrounding who
copes and who chokes under pressure (Hill et al., 2010; Otten, 2009).
Attentional Control Theory: Sport (ACTS; Eysenck & Wilson, 2016), aims to address this
gap by not only describing the effect of anxiety on performance, but by considering how anxiety
arises in the first place. ACTS maintains the predictions of Attentional Control Theory (ACT;
Eysenck et al., 2007) which proposes that anxiety causes increased attention to threat-related cues as a
result of a disrupted balance between top-down and bottom-up attentional systems (Cocks et al., 2016;
Corbetta & Shulman, 2002; Wilson, 2008). However, ACTS extends this core mechanism to suggest
that the origins of competitive anxiety are rooted in ongoing appraisals of the costs of failure (‘what’s
at stake?’) and the probability of failure (‘how am I doing?’) (Berenbaum, 2010; Martens et al., 1990;
see Figure 1). While external factors (e.g., social comparison, monetary reward) might create
situational pressure, it is only the appraisal of whether success is important and whether failure is
likely, that will lead to the initiation of the experience of anxiety. If the performer sees the cost of
failure as low, or unlikely to occur, they may well avoid the anxiety-inducing effect of psychological
pressure.
In addition to describing the appraisals that precipitate anxiety, ACTS identifies the important
role that momentary errors may play in this process, an issue that has received limited attention within
sport psychology. Within the ACTS framework (see Figure 1), the perceived probability of future
failure increases as a function of the number of recent failure experiences; primarily mental and
physical errors (e.g., Nicholls et al., 2005). Crucially, ACTS predicts an interactive effect, whereby
errors are more likely to be attended to and interpreted more negatively when anxiety is already high
(i.e., increased attention to threat; Eysenck et al., 2007). Support for this postulate comes from
findings in both mainstream cognitive psychology and cognitive neuroscience. For instance, a series
of four experiments by Liu, Shen, and Li (2019) using a dot-probe task demonstrated a positive
feedback loop between state anxiety and attentional bias, in which state anxiety directly increased
attentional bias towards negative words and an experimentally-induced negative attentional bias
increased state anxiety under stressful conditions. This effect of attentional bias on state anxiety (but
not the reverse effect) was found to be moderated by cognitive appraisals (see also Basanovic et al.,
2020). Additionally, Aarts and Pourtois (2012) found that an electroencephalogram (EEG) event-
related potential sensitive to the valence of feedback indicated that error monitoring was more
Accepted for publication in Psychology of Sport and Exercise 25/05/21 PREPRINT
3
disrupted in highly anxious individuals. Consequently, the co-occurrence of high levels of
psychological pressure and prior performance errors may be a recipe for choking, if individuals are
more attuned to failure cues (errors) and appraise them as likely to influence subsequent performance.
A recent study supported this error dependency effect in elite sport (American Football),
finding that when one error was made, the probability of making another on the subsequent play
increased (Harris et al., 2019). Harris et al. examined all plays from the National Football League
(NFL) over seven seasons, using a scoring system to identify the occurrence of high-pressure plays
and substantial errors. As has been found previously in laboratory-based (Cooke et al., 2010) and real-
world (Hickman & Metz, 2015; Pocock et al., 2018; Toma, 2017) studies, an increase in performance
pressure resulted in more frequent errors. Crucially for the predictions of ACTS, the detrimental
effects of psychological pressure and a preceding error also had an interactive effect, causing the
greatest disruption when a play was at a crucial moment of the game and preceded by an error.
Indeed, the probability that one error would follow another almost doubled from 27% on low pressure
plays to 50% on high pressure plays (Harris et al., 2019).
The findings of Harris et al. (2019) did not, however, permit examination of individual
differences and trait level factors in coping strategies and choking responses that might moderate the
effects of situational pressure. One potentially important trait discussed in the sport psychology
literature is that of habitual ‘clutch’ performers (Hibbs, 2010; Otten, 2009; Schweickle et al., 2021).
Clutch performers are individuals who customarily respond well to psychological pressure by
avoiding choking and even improving their performances (Schweickle et al., 2021; Swann et al.,
2017). However, real-world game statistics have not convincingly supported the idea that particular
individuals consistently raise their performances under pressure (Birnbaum, 2008). For instance,
Solomonov et al. (2015) showed that NBA (National Basketball Association) players commonly
thought to be ‘clutch players’ did not actually improve their shooting percentage in the last 5 minutes
of games. However, they did take more shots which may account for the perception of ‘clutch’. As
such, the existence of habitual positive responders to pressure remains uncertain.
Accepted for publication in Psychology of Sport and Exercise 25/05/21 PREPRINT
4
Figure 1.
Schematic representation of the bi-directional pressure-performance relationship, as outlined in Attentional
Control Theory (Eysenck et al., 2007; dark grey; bottom right) and Attentional Control Theory: Sport (Eysenck
& Wilson, 2016; light grey; top left). Dashed lines indicate a weaker effect. The top left of the model describes
how psychological pressure does not lead directly to anxiety but is mediated by cognitive appraisals of cost and
probability of failure. The top right of the model depicts a feedback loop from prior performance, illustrating
the effect that errors may have on cognitive appraisals.
The Present Study
Examining large, real-world data sets (e.g., as in Harris et al., 2019; Hickman & Metz, 2015;
Toma, 2017) offers greater statistical power and opportunities to test predictions about pressure
outside of the artificial laboratory environment, where ‘blocked’ experimental conditions are unlikely
to reveal anything about the drivers of anxiety and its effect on the critical occasions when
performance truly matters. For instance, Toma (2017) found evidence of widespread choking in both
men’s and women’s college and professional basketball, with players less likely to successfully
execute free throws in the final minutes of close games. Similarly, Hickman and Metz (2015) found
that for individual putts taken on the PGA tour, as the amount of money riding on the shot increased
(i.e., potential change in earnings), so did the likelihood of a miss. The adoption of existing real-world
data sets in the current study allowed us to test the overarching predictions of ACTS (Eysenck &
Wilson, 2016) in relation to pressure, errors, and performance outcomes in the real-world. It is
recognised, however, that it is not possible to assess the intervening cognitive mechanisms with this
approach (cf. Aarts & Pourtois, 2012; Liu et al., 2019). Therefore, we do not offer a full test of ACTS,
but rather use the performance-related predictions as a basis for examining the effects of pressure and
prior errors.
Accepted for publication in Psychology of Sport and Exercise 25/05/21 PREPRINT
5
The aim of the study was to replicate and extend the findings of Harris et al. (2019) in relation
to the play-by-play effects of pressure and prior errors on performance in American Football to a
point-by-point perspective in Grand Slam tennis tournaments. Grand Slam tournaments were chosen
as they represent the most prestigious and lucrative tournaments for elite players, and hence are likely
to elicit maximal psychological pressure. The results of Harris et al. were complicated by the nature of
team sports (e.g., in how pressure is differentially experienced, and errors attributed). Consequently,
we aimed to replicate these findings in an individual sporting endeavour (singles tennis) where
unforced errors (rather than good plays by an opponent) could be more easily determined. Based on
the predictions of ACTS, and prior findings in American Football (Harris et al., 2019), it was
hypothesised that pressure would have a negative effect on performance (i.e., more unforced errors)
(H1); prior errors would have a negative effect on performance (H2); and there would be an interactive
effect, such that the effect of a preceding error would be greatest under higher pressure (H3). It was
also predicted that changes in the rate of unforced errors would be a result of impaired skill execution
(choking) rather than a general change in playing strategy (i.e., the balance of risk/reward in shot
selection). Therefore, it was expected that changes in unforced errors as a result of pressure or prior
errors would not be accompanied by a similar increase in ‘winners’ (as a measure of higher-risk shots)
(H4).
To extend previous work using real-world data sets, we also aimed to examine differential
responses to psychological pressure and errors to determine whether players who were subsequently
successful coped better with pressure and/or responded better to in-game mistakes. Based on real-
world data it has previously been suggested that highly-ranked professional tennis players perform
particularly well when the stakes are high, as more highly-ranked players beat lower ranked players
more frequently when playing in Grand Slam tournaments (Jetter & Walker, 2015). However, the sum
of the evidence for ‘clutch’ performers is weak (Birnbaum, 2008; Solomonov et al., 2015). Therefore,
we tentatively hypothesised that all players would be affected by pressure and prior errors, and that
any advantage for subsequently successful players at higher levels of pressure and after errors would
be commensurate with their general performance advantage (H5).
Methods
We performed a retrospective analysis of existing data from the four tennis Grand Slam
tournaments (Australian Open, French Open, Wimbledon, and US Open). Point-by-point data from
grand slam matches (all men’s and women’s singles matches) between 2016-2019 were scraped from
the tournament websites, corresponding to 12 tournaments, 3,552 matches and 658,068 individual
points of tennis
1
. Four tournaments the French Open and Australian Open for 2018 and 2019 had
some missing data (the occurrence of double faults and unforced errors was not recorded) so these
1
https://github.com/JeffSackmann/tennis_slam_pointbypoint
Accepted for publication in Psychology of Sport and Exercise 25/05/21 PREPRINT
6
tournaments were excluded from analysis. The relevant variables extracted from the dataset were: the
ongoing score in the match (sets and games); which player was serving; which player won the point;
and when break points, double faults, unforced errors and clean winners occurred. The full data set is
available online and the data included in the analysis, as well as the analysis code, is available from
the Open Science Framework (https://osf.io/b4vpd/).
Measures
Errors
Errors were operationalised as a double fault or an unforced error, as instances of mistakes
that were controllable, and not the direct result of a good shot from the opponent. Both unforced
errors and double faults were already coded in the data from the Grand Slam websites and were
combined into an overall ‘unforced errors’ performance measure. Next, points on which the player
had made an unforced error on the immediately preceding point were coded as ‘post-error’ points
(i.e., on the previous point within the same game).
Winners
In tennis, winners are shots that are unreturned by the opposing player and which they do not
get their racquet on. Consequently, a winner is not only an instance of successful shot execution, but
potentially also a higher risk shot than one hit to where the opponent can reach the ball.
Pressure
Scoring of psychological pressure was based on a system devised by the authors but derived
from similar work (Deutscher et al., 2018; Harris et al., 2019; Hickman & Metz, 2015; Toma, 2017).
From a theoretical perspective, the occurrence of pressure is a result of conditions that increase the
importance of performing well (Baumeister, 1984), such as playing in the final of a Grand Slam
tournament. Therefore, points that were played towards the end of games, towards the end of sets, and
towards the end of matches were deemed to be higher pressure as they had a more direct effect on the
outcome of the contest. Additionally, game points and break points (where winning the point would
result in winning the game for one of the players) were deemed to add additional pressure, because
the importance of performing well was inherently higher than earlier points in a game. Consequently,
pressure was assigned in a cumulative manner whereby all points started with a base score of 1, which
was increased when any of the following occurred:
1) It was late in the game (i.e., both players on 30/40/adv);
2) If the opponent could win the set in the game (so 5-4 or 5-1 or 6-5 etc.);
3) The match had gone to a deciding set (5th for men, 3rd for women);
4) If opponent had game point;
5) If facing break point.
Accepted for publication in Psychology of Sport and Exercise 25/05/21 PREPRINT
7
Therefore, if a player was 40-30 down on their own serve, in a deciding set, and down 5 games to 4,
this would create the maximum pressure score. During data analysis it was observed that the
maximum score occurred very infrequently (only when facing a break point coinciding with match
point), so the last two pressure categories (scores of 5 or 6) were combined to provide a pressure
index ranging from 1 to 5, with 5 indicating highest pressure (see Harris et al., 2019).
Data Analysis
Linear mixed effects models (LMMs) were used to examine the effect of pressure and prior
errors on performance, using the lme4 package for R (Bates et al., 2014). The use of a mixed effects
model reflects the assumption that there is likely a range of true effects across participants, matches,
and tournaments, from which we aim to estimate the mean. Initially, near maximal models were
fitted, using random factors for participant (slope and intercept) and participant nested within
tournament (slopes and intercepts) (Barr et al., 2013). The models were then simplified to provide the
most parsimonious fit based on Principal Components Analysis, using the RePsychLing package, as
described by Bates et al. (2018), and the Akaike information criterion for comparing competing
models. We report an odds ratio (OR) as an effect size for the dichotomous outcome variables
unforced errors and winners. All analysis scripts and raw data are available from the Open Science
Framework (https://osf.io/b4vpd/).
Results
Errors
A linear mixed effects model (with player as a random factor) was run to examine the effect
of game pressure and prior errors on performance. Firstly, a prior unforced error was found to
increase the chance of a further unforced error (OR=1.2, 95%CI [1.17, 1.23], p<.001). Secondly, an
increase in pressure to a score of 3, 4, or 5 was also found to increase the rate of unforced errors
(ps<.001). Pairwise comparisons, with a Bonferroni-Holm adjustment, indicated the rate of errors at
each level of pressure to be significantly different from all others at p<.001, apart from levels of 4 and
5 which were only marginally different (p=.05), and levels of 2 and 1 which were not significantly
different (p=.80) (see Figure 3 Panel A).
Significant interaction effects at pressure levels of 2 (p=.03), 3 (p=.008) and 4 (p<.001) were
also found, which indicated that the effect of a prior error was exacerbated as pressure increased.
Pairwise tests with Bonferroni-Holm correction indicated that the likelihood of an error was
significantly higher on post-error points for all levels of pressure score (ps<.001) (see Figure 2 Panel
A and Figure 3 Panel B).
Accepted for publication in Psychology of Sport and Exercise 25/05/21 PREPRINT
8
Table 1
Odds ratios (and their 95% CIs) for the fixed effects split by level
OR
CI low
(2.5%)
p-value
Intercept
0.18
0.18
<.0001***
Pressure 2
1.01
0.99
.299
Pressure 3
1.14
1.12
<.0001***
Pressure 4
1.47
1.41
<.0001***
Pressure 5
1.86
1.59
<.0001***
Post error
1.20
1.18
<.0001***
Pressure 2 * Post error
1.04
1.01
.025*
Pressure 3 * Post error
1.07
1.02
.008**
Pressure 4 * Post error
1.24
1.13
<.001***
Pressure 5 * Post error
1.33
0.93
.12
Note: reference categories for the odds ratios are a pressure score of 1 and no error. *p<.05, **p<.01,
***p<.001
Figure 2
Plotted point estimate (and 95% CIs) for effect of model predictors on unforced errors (Panel A) and
winners (Panel B).
Note: Reference categories for the odds ratios are a pressure score of 1 and no error
Winners
To examine whether changes in the rate of unforced errors across levels of pressure was
confounded by a change in playing strategy (i.e., were players just less conservative in their play
following an error when under pressure?) we also examined the rate at which players hit winners (i.e.,
unreturnable shots). A linear mixed effects model (with player as a random factor) indicated that
winners were less frequent at pressure scores of 2 (OR=0.92, 95%CI [0.91, 0.94], p<.001) or 3
(OR=0.96, 95%CI [0.94, 0.99], p=.001) compared to the reference category (index of 1), but there
were no significant effects at other levels (ps>.22) and no consistent pattern of effect (see Figure 2
0.0 0.5 1.0 1.5 2.0 2.5
Pressure 5 * Post Error
Pressure 4 * Post Error
Pressure 3 * Post Error
Pressure 2 * Post Error
Post Error
Pressure 5
Pressure 4
Pressure 3
Pressure 2
Intercept
Odds ratio
Estimated ORs for model predictors: Errors
0.0 0.5 1.0 1.5 2.0
Pressure 5 * Post Error
Pressure 4 * Post Error
Pressure 3 * Post Error
Pressure 2 * Post Error
Post Error
Pressure 5
Pressure 4
Pressure 3
Pressure 2
Intercept
Odds ratio
Estimated ORs for model predictors: Winners
A
B
Accepted for publication in Psychology of Sport and Exercise 25/05/21 PREPRINT
9
Panel B and Figure 3 Panel B). This result confirms that changes in error rates under pressure could
not be explained by a change to a more high-risk strategy. Winners were, however, found to be more
frequent on points following an error (OR=1.10, 95%CI [1.08, 1.13], p<.001) (see Figure 2 Panel B).
Small interaction effects were found at pressure levels of 2 (OR=1.08, 95%CI [1.05, 1.12], p<.001)
and 3 (OR=1.22, 95%CI [1.16, 1.28], p<.001), but not at 4 or 5 (ps>.97). Pairwise tests with
Bonferroni-Holm correction indicated that the likelihood of a winner was significantly higher on post-
error points at low levels of pressure (1, 2, and 3; ps<.005), but not higher levels (4 and 5; ps>.12).
Figure 3
Probability of unforced errors (mean and 95% CIs) across pressure score (Panel A) and probability of unforced
errors (bars) and winners (points) across pressure score, split by preceding error (Panel B).
Individual Responses to Pressure and Errors
To explore potential differential effects in responding to psychological pressure and prior
errors, we tested whether there was a clear clustering of datapoints (see Figure 4) representing high
rates of winners and low rates of unforced errors for players who subsequently won their match (i.e.,
successful), compared to those who lost (i.e., unsuccessful). An algorithm called the interpoint
distance (IPD) test (see Marozzi et al., 2020) was used to determine whether the underlying
distributions of errors/winners among successful and unsuccessful players were statistically different.
IPD testing enables comparisons of the mean, variance, and underlying distributions of high-
dimensional data based on the Euclidean distance between points, free from any assumptions about
underlying distributions. Additionally, a 95% confidence interval (CI) bounding ellipse of the data for
successful and unsuccessful players was calculated, and the degree of overlap between the two
ellipses was determined using the maximum likelihood method. For all points across all games
(Figure 4 Panel A) the IPD test indicated that the underlying distributions of winners to unforced
errors for successful versus unsuccessful players were significantly different (p<.001), with 54.8%
0.0 0.1 0.2 0.3 0.4
1
2
3
4
5
Probability
Pressure score
Unforced error: No prior error
Unforced error: Prior error
Effect of pressure and errors on unforced errors
and winners
Winners: No prior error
Winners: Prior error
0.0 0.1 0.2 0.3 0.4
1
2
3
4
5
Probability
Pressure score
Effect of pressure on unforced errors
AB
Accepted for publication in Psychology of Sport and Exercise 25/05/21 PREPRINT
10
overlap in the 95% CI ellipses. For higher-pressure points (index of 3+; Figure 4 Panel B),
distributions were again significantly different (p<.001) with a similar level of overlap (55.8%). For
points following an error (Figure 4 Panel C) the IPD test indicated that distributions were again
significantly different (p<.001) but showed a greater degree of overlap (67.5%). Finally, for high
pressure points that also followed an error (Figure 4 Panel D), the distributions remained significantly
different (p<.001), but showed a further increase in overlap (76.8%). Consequently, there was no
evidence for differential effects of pressure and errors: winning players generally had a better ratio of
winners to unforced errors, but this difference did not change for high pressure points and was
actually reduced for points after an error, particularly when coinciding with high pressure.
Accepted for publication in Psychology of Sport and Exercise 25/05/21 PREPRINT
11
Figure 4
Jittered scatter plots (with marginal distribution histogram) of the rate of unforced errors against the rate of
winners for all points in all games (A), for the higher pressure (3+ score) points (B) for the post error points
(C), and for the post error points coinciding with high pressure (D) colour coded by match winner. 95%
confidence interval error ellipses, and the percentage overlap between winners/losers have been calculated.
Discussion
Direct testing of predictions derived from theory is important for refining and developing
better theoretical models in psychology (Ferguson & Heene, 2012). In the context of elite-level tennis,
we sought to examine the effects of psychological pressure and prior errors on subsequent
performance. In particular, we aimed to test the overarching predictions of ACTS (Eysenck & Wilson,
2016) that pressure and prior errors would have multiplicative effects on performance. In line with our
first hypothesis, there was a clear effect of pressure, with more unforced errors occurring as the
Accepted for publication in Psychology of Sport and Exercise 25/05/21 PREPRINT
12
pressure index increased. There was little difference in error rate between pressure scores of 1 and 2,
but all subsequent increases in pressure resulted in a corresponding increase in errors (see Figure 3
Panel A). The unforced error rate for the highest-pressure points (0.28±0.45) was 1.75 times that of
low-pressure points (0.16±0.36). The observed effect of pressure on performance is in line with prior
work examining real-world data sets in basketball (Toma, 2017) and golf (Hickman & Metz, 2015),
although a contrasting finding has been reported by Deutscher et al. (2018) in darts.
Deutscher et al. (2018) suggested that high-level darts players actually perform better under
pressure, as evidenced by improved performance when faced with more challenging game situations:
a three-dart finish as opposed to a one-dart finish (but with three opportunities) when the opponent
can also finish on their next visit. However, Deutscher et al. interpreted pressure solely as a function
of absolute probability of successful outcome, ignoring the role that cost of failure plays in the
perception of pressure (Berenbaum, 2010). While hitting three specific shots to finish is clearly much
harder than only hitting one, the expectation of success is very different. One of the major costs for a
sporting performer is the effect that failure has on their own self-concept/self-esteem (referred to as
‘ego threat’; Leary et al., 2009), therefore it could be argued that one-dart finishes are likely to be
more pressurised than three-dart finishes given the increased expectation of success. Considered in
this light, professional darts players may not in fact be excelling under pressure, but choking, as has
been shown in basketball (Toma, 2017), golf (Hickman & Metz, 2015), and American Football
(Harris et al., 2019).
Our second hypothesis concerned the effect of preceding errors on subsequent performance
and the idea that there may be a degree of dependence between successive points. The results
indicated that the rate of unforced errors was significantly increased on points following an error and
hence supported the potentially detrimental error feedback effect (Aarts & Pourtois, 2012; Baumeister
et al., 2001; Eysenck & Wilson, 2016; Harris et al., 2019). Our third hypothesis, predicting an
interactive effect between pressure and errors, was also supported. At pressure index scores of 2, 3,
and 4 a significant interaction effect, over and above the additive effects of pressure and errors, was
observed. This interaction was not detected at a pressure score of 5 (p=.12), however the subset of
data at the highest level of pressure was much sparser than other levels. Figure 3 (Panel B) reflects
that the interaction pattern appears to continue to the pressure index of 5, but with a much less certain
point estimate, which may explain the lack of a significant effect.
The observed interactive effect of pressure and errors is in line with the predictions outlined
in ACTS where the perceived probability of failure is likely to increase after an error (i.e.,
dependence) and interacts with the cost of failure (which is greater at critical points in a match) to
raise performance anxiety. Previous investigations of the human performance-monitoring system have
demonstrated how error detection serves as a signal to adapt and improve ongoing behaviour
(Botvinick et al., 2001). Yet, in the context of high anxiety this process may prove maladaptive.
Accepted for publication in Psychology of Sport and Exercise 25/05/21 PREPRINT
13
Masaki et al. (2017) have previously shown how sports anxious individuals exhibit neural markers of
abnormal error monitoring when performance was evaluated. While the appraisal of errors could not
be measured in the present work, when considered together with previous findings that anxiety
increases attention to threat (Bar-Haim et al., 2007), increases error monitoring (Moser et al., 2013),
and leads to difficulty disengaging with emotionally negative information (Fox et al., 2001), the co-
occurrence of high-pressure and a prior error may set off a chain of events where negative stimuli
(e.g., previous failure and the prospect of future failure) are fixated on, causing further anxiety and
further mistakes. This sequencing of anxiety, error monitoring, and attention to threat should be
further investigated in future research on performance under pressure.
While we were unable to affirm the intervening cognitive mechanisms relating to attentional
bias and cognitive appraisals outlined in ACTS, the observed interaction effect cannot simply be
explained by a shift towards a more high-risk strategy (i.e., more unforced errors and winners) under
pressure. Our analysis of ‘winners’ did not reveal any effect of pressure nor an interaction effect that
would indicate a high-risk strategy in response to pressure. Indeed, the opposite a more conservative
approach to avoid further errors would seem more plausible and future work may wish to examine
potential in-game changes in strategy, particularly in relation to the cognitive appraisal of prior
mistakes and their perceived probability and costs (Berenbaum, 2010). Consequently, the effect of
pressure on unforced errors is most likely related to a skill breakdown, as described in theories of
anxiety and pressure on sporting performance (Beilock & Carr, 2001; Eysenck & Wilson, 2016;
Masters & Maxwell, 2008; Nieuwenhuys & Oudejans, 2012; Vine et al., 2016; Masters, 1992;
Nieuwenhuys & Oudejans, 2012). Future work may wish to examine the cognitive appraisal of errors
in more detail by collecting self-reported appraisals after experimentally manipulated error feedback.
This would enable a more comprehensive test of the predictions of ACTS (see Figure 1) in terms of
how the appraisal of errors relates to changes in attention, and subsequently performance.
To examine whether the observed pressure and error effects were universal, or whether
subsequently successful players were better able to avoid additional unforced errors, we examined the
distributions of winners and unforced errors using an interpoint distance test (H5). Research into
‘clutch’ performance has suggested that some individuals cope better under psychological pressure
(Otten, 2009), but our findings provided no support for this effect. Predictably, successful players
were found to have a much better ratio of winners to unforced errors across all points (Figure 4 Panel
A). However, for higher pressure points the distributions of winners to unforced errors for
successful/unsuccessful players (Figure 4 Panel B) was very similar to the overall distributions,
indicating that successful players simply maintained their general advantage under pressure and did
not somehow ‘raise their game’. On post-error points (Figure 4 Panel C), the distinction between
successful and unsuccessful players was much reduced, suggesting that all players struggled to
recover from errors. Further, on post-error points that were also higher pressure, the distinction
Accepted for publication in Psychology of Sport and Exercise 25/05/21 PREPRINT
14
between successful and unsuccessful players was smaller still, again suggesting that the co-occurrence
of pressure and a prior error had a damaging effect, regardless of the player’s ability. Clearly,
individualised responses to sporting situations will determine whether performers perceive any
pressure and how they will respond in performance terms (Schweickle et al., 2021), but as our data
was anonymised, we could not explore potentially interesting individual factors (e.g., player ranking
or recent performance history). However, these findings provided no evidence for any widespread
‘clutch’ effect in winning players, and instead supported the ubiquity of the pressure and error
feedback effects.
It is important to note that our findings contrast with previous work which has reported
higher-ranked professional tennis players to improve their winning percentage in clutch situations
(Jetter & Walker, 2015). However, the strongest evidence for clutch effects reported by Jetter and
Walker was that higher-ranked players increased their probability of beating lower ranked players
when playing in Grand Slam tournaments. An alternative interpretation of Jetter and Walker’s data is,
however, that highly ranked players simply under-performed in less important tournaments, where
their motivation may have been lower. Further, competing in best-of-3 set matches, as opposed to 5
(for men) in non-grand slam tournaments increases the odds of extreme results. Jetter and Walker do
report that the evidence for more acute within-game instances of clutch performance that higher
ranked players would excel in tie breaks and deciding sets was weak in comparison to the broader
effect of playing in Grand Slam events. Additionally, Jetter and Walker were able to identify the top
ranked players in the world which was not possible in the current data set. The clutch abilities of these
super-elite players may be different to the broader successful / unsuccessful effects which we report
here.
Nonetheless, our findings do question the idea that subsequently successful players (on
average better players) responded more positively to pressure, and instead demonstrate that the effects
of situational pressure and prior mistakes affect even the most elite athletes. Indeed, the weight of
evidence is strongly in favour of an overall performance degradation under pressure (Harris et al.,
2019; Hickman & Metz, 2015; Toma, 2017), and the notion of ‘clutch players’ may be a cognitive
bias present in observers rather than a real effect (Solomonov et al., 2015). Individuals commonly
thought to be ‘clutch’ players Michael Jordan, Tom Brady, Wayne Gretsky might well perform at
important moments simply because of their general superiority, not because they are immune to the
effects of pressure.
One of the practical implications of the present findings is a greater appreciation of the
detrimental impact prior errors can have and the need for athletes to develop strategies to mitigate
against error dependency effects. While it might be worthwhile to allow performers to commit (and
then correct) errors as they are learning skills (Metcalfe, 2017), the fact that anxious apprehension is
associated with exaggerated error monitoring (Moser et al., 2013) means that the perception of errors
Accepted for publication in Psychology of Sport and Exercise 25/05/21 PREPRINT
15
can change in high stakes environments. As outlined in ACTS, the cognitive appraisal of an error
determines the subsequent effect on anxiety and performance, via its influence on the perceptions of
the probability of failure (Berenbaum, 2010). Therefore, developing the ability to respond more
positively to errors may be crucial in breaking the feedback loop between attentional bias to threat
cues, state anxiety, and cognitive appraisals of threat (Eysenck & Wilson, 2016; Liu et al., 2019).
Instances of athletes choking under pressure are often linked to an initial error that induces anxiety,
overthinking, and a detrimental focus of attention on skill execution (Hill et al., 2010; Williams &
Wigmore, 2020). While the present dataset indicated that even winning players struggled in the wake
of an error, the ability to simply forget these errors might be a highly valuable mental skill.
This type of post-error recovery is even quantified for the PGA tour as a ‘bounce back’
percentage
2
; reflecting the frequency by which a player follows up an over par score on a hole with an
under-par score on the subsequent hole. By way of a real-world example, Annika Sörenstam (the 3rd
most successful golfer of all time on the Ladies PGA tour) reports that she barely remembers making
an error; after a bad shot she would conduct a brief analysis of the mistake, then move on and focus
on the ‘now shot’ (Williams & Wigmore, 2020). Incidentally, the greater length of time between
shots in golf compared to tennis may make forgetting easier, and therefore sport type may be an
important moderator of this relationship. Frameworks like ACTS can potentially provide a basis for
developing practical techniques to avoid negative appraisals of errors, which can otherwise influence
fragile performance states like sport confidence (Beaumont et al., 2015). For instance, techniques
such as pre-shot routines and constructive self-talk may help to develop this form of ‘error amnesia’
(Harris et al., 2019), and robust sport confidence (Thomas et al., 2011) as described by Sörenstam.
The benefits of pre-shot routines for performing under pressure are varied and not fully
understood (e.g., see Cotterill, 2010; Mesagno & Mullane-Grant, 2010). However, recent research in
cognitive neuroscience suggests that even arbitrary rituals can dull the neural response to performance
failure and buffer against uncertainty and anxiety (Hobson et al., 2017). In the language of ACTS, this
means that as the cost of failure increases (as pressure is raised) pre-shot routines may disrupt the
error feedback cycle that strongly influences perceptions of failure probability. However, as routines
are more easily applicable to self-paced skills (e.g., the tennis serve in our data), self-talk
interventions, focusing on directing attentional control ‘in the moment’, may be more appropriate for
skills when time constraints are more pronounced (e.g., rally strokes and volleys in our data). Again,
the self-regulatory benefits of intentionally using self-talk are numerous; including directing
attentional focus, enhancing confidence, regulating effort, and controlling emotional and cognitive
reactions (Latinjak et al., 2019; Theodorakis et al., 2008). In the specific context of reducing attention
towards potentially threatening failure signals, there is some evidence that self-talk interventions
2
https://www.pgatour.com/stats/stat.160.html
Accepted for publication in Psychology of Sport and Exercise 25/05/21 PREPRINT
16
might be useful. Schüler and Langens (2007) tested the use of self-talk strategies as a means for
buffering against the negative effects of psychological crisis (‘hitting the wall’) in non-professional
runners completing a marathon. They reported that among runners who experienced hitting the wall,
those using self-talk coped better than those in a control group. Consequently, self-talk may be an
effective intervention for both regulating responses to pressure and mitigating against the negative
feedback cycle of performance errors (Cooper et al., 2020).
Conclusions
In summary, the present study aimed to examine the negative effects of situational pressure
and prior errors on subsequent performance in elite tennis. The results replicate those reported by
Harris et al. (2019) in American Football; that both situational pressure and prior errors interact to
induce further performance breakdowns. The present findings extend this work to a new sport that
avoids the complication of team performance. The findings further clarify the likely momentary
conditions where performance might be disrupted, in even the most skilled of athletes, although it is
possible the few top players in the world do not show these effects. In particular, the results speak to
the power of pressure and prior errors, and the feedback loop that they may form with ongoing
cognitive appraisals of the probability and costs of success and failure. However, while the broader
performance feedback effects postulated in ACTS are supported, we cannot speak to the intervening
mechanisms here. Therefore, future work should attempt the difficult task of examining how moment-
to-moment fluctuations in meaningful pressure affects cognitive appraisals of the probability of
success and failure, and how these appraisals relate to changes in anxiety, attention, and performance.
Accepted for publication in Psychology of Sport and Exercise 25/05/21 PREPRINT
17
References
Aarts, K., & Pourtois, G. (2012). Anxiety disrupts the evaluative component of performance
monitoring: An ERP study. Neuropsychologia, 50(7), 12861296.
https://doi.org/10.1016/j.neuropsychologia.2012.02.012
Bar-Haim, Y., Lamy, D., Pergamin, L., Bakermans-Kranenburg, M. J., & van IJzendoorn, M. H.
(2007). Threat-related attentional bias in anxious and nonanxious individuals: A meta-analytic
study. Psychological Bulletin, 133(1), 124. https://doi.org/10.1037/0033-2909.133.1.1
Barr, D. J., Levy, R., Scheepers, C., & Tily, H. J. (2013). Random effects structure for confirmatory
hypothesis testing: Keep it maximal. Journal of Memory and Language, 68(3), 255278.
https://doi.org/10.1016/j.jml.2012.11.001
Basanovic, J., Kaiko, I., & MacLeod, C. (2020). Change in Attentional Control Predicts Change in
Attentional Bias to Negative Information in Response to Elevated State Anxiety. Cognitive
Therapy and Research. https://doi.org/10.1007/s10608-020-10176-3
Bates, D., Kliegl, R., Vasishth, S., & Baayen, H. (2018). Parsimonious Mixed Models.
ArXiv:1506.04967 [Stat]. http://arxiv.org/abs/1506.04967
Bates, D., Mächler, M., Bolker, B., & Walker, S. (2014). Fitting linear mixed-effects models using
lme4. ArXiv:1406.5823 [Stat]. http://arxiv.org/abs/1406.5823
Baumeister, R. F. (1984). Choking under pressure: Self-consciousness and paradoxical effects of
incentives on skillful performance. Journal of Personality and Social Psychology, 46(3), 610
620.
Baumeister, R. F., Bratslavsky, E., Finkenauer, C., & Vohs, K. D. (2001). Bad Is Stronger Than Good.
Review of General Psychology, 5(4), 323370. https://doi.org/10.1037/1089-2680.5.4.323
Baumeister, R. F., & Showers, C. J. (1986). A review of paradoxical performance effects: Choking
under pressure in sports and mental tests. European Journal of Social Psychology, 16(4), 361
383. https://doi.org/10.1002/ejsp.2420160405
Beaumont, C., Maynard, I. W., & Butt, J. (2015). Effective Ways to Develop and Maintain Robust
Sport-Confidence: Strategies Advocated by Sport Psychology Consultants. Journal of Applied
Sport Psychology, 27(3), 301318. https://doi.org/10.1080/10413200.2014.996302
Beilock, S. L., & Carr, T. H. (2001). On the fragility of skilled performance: What governs choking
under pressure. Journal of Experimental Psychology: General, 130, 701725.
Beilock, S. L., & Gray, R. (2007). Why do athletes choke under pressure? In Handbook of sport
psychology, 3rd ed (pp. 425444). John Wiley & Sons, Inc.
Berenbaum, H. (2010). An initiationtermination two-phase model of worrying. Clinical Psychology
Review, 30(8), 962975. https://doi.org/10.1016/j.cpr.2010.06.011
Birnbaum, P. (2008). Clutch hitting and the cramer test. Baseball Research Journal, 37, 7175.
Botvinick, M. M., Braver, T. S., Barch, D. M., Carter, C. S., & Cohen, J. D. (2001). Conflict
monitoring and cognitive control. Psychological Review, 108(3), 624652.
https://doi.org/10.1037/0033-295X.108.3.624
Cocks, A. J., Jackson, R. C., Bishop, D. T., & Williams, A. M. (2016). Anxiety, anticipation and
contextual information: A test of attentional control theory. Cognition and Emotion, 30(6),
10371048. https://doi.org/10.1080/02699931.2015.1044424
Cooke, A., Kavussanu, M., McIntyre, D., & Ring, C. (2010). Psychological, muscular and kinematic
factors mediate performance under pressure. Psychophysiology, 47(6), 11091118.
https://doi.org/10.1111/j.1469-8986.2010.01021.x
Accepted for publication in Psychology of Sport and Exercise 25/05/21 PREPRINT
18
Cooper, K. B., Wilson, M. R., & Jones, M. I. (2020). Fast talkers? Investigating the influence of self-
talk on mental toughness and finish times in 800-meter runners. Journal of Applied Sport
Psychology, 0(0), 119. https://doi.org/10.1080/10413200.2020.1735574
Corbetta, M., & Shulman, G. L. (2002). Control of Goal-Directed and Stimulus-Driven Attention in the
Brain. Nature Reviews Neuroscience, 3(3), 201215.
Cotterill, S. (2010). Pre-performance routines in sport: Current understanding and future directions.
International Review of Sport and Exercise Psychology, 3(2), 132153.
https://doi.org/10.1080/1750984X.2010.488269
Deutscher, C., Ötting, M., Langrock, R., Gehrmann, S., Schneemann, S., & Scholten, H. (2018). Very
Highly Skilled Individuals Do Not Choke Under Pressure: Evidence from Professional Darts.
ArXiv:1809.07659 [Stat]. http://arxiv.org/abs/1809.07659
Eysenck, M. W., Derakshan, N., Santos, R., & Calvo, M. G. (2007). Anxiety and cognitive
performance: Attentional control theory. Emotion, 7(2), 336353. https://doi.org/10.1037/1528-
3542.7.2.336
Eysenck, Michael W. (2013). Anxiety: The Cognitive Perspective. Psychology Press.
Eysenck, M.W., & Wilson, M. R. (2016). Sporting performance, pressure and cognition: Introducing
attentional control theory: Sport. In D. Groome & M. Eysenck (2nd ed). An introduction to
applied cognitive psychology. (Pp. 329-350). London: Routledge.
Ferguson, C. J., & Heene, M. (2012). A Vast Graveyard of Undead Theories: Publication Bias and
Psychological Science’s Aversion to the Null. Perspectives on Psychological Science, 7(6),
555561. https://doi.org/10.1177/1745691612459059
Fox, E., Russo, R., Bowles, R., & Dutton, K. (2001). Do threatening stimuli draw or hold visual
attention in subclinical anxiety? Journal of Experimental Psychology: General, 130(4), 681
700. https://doi.org/10.1037/0096-3445.130.4.681
Harris, D. J., Vine, S. J., Eysenck, M. W., & Wilson, M. R. (2019). To err again is human: Exploring a
bidirectional relationship between pressure and performance failure feedback. Anxiety, Stress,
& Coping, 32(6), 670678. https://doi.org/10.1080/10615806.2019.1643459
Hibbs, D. (2010). A Conceptual Analysis of Clutch Performances in Competitive Sports. Journal of the
Philosophy of Sport, 37(1), 4759. https://doi.org/10.1080/00948705.2010.9714765
Hickman, D. C., & Metz, N. E. (2015). The impact of pressure on performance: Evidence from the
PGA TOUR. Journal of Economic Behavior & Organization, 116, 319330.
https://doi.org/10.1016/j.jebo.2015.04.007
Hill, D. M., Hanton, S., Matthews, N., & Fleming, S. (2010). Choking in sport: A review. International
Review of Sport and Exercise Psychology, 3(1), 2439.
https://doi.org/10.1080/17509840903301199
Hobson, N. M., Bonk, D., & Inzlicht, M. (2017). Rituals decrease the neural response to performance
failure. PeerJ, 5. https://doi.org/10.7717/peerj.3363
Janelle, C. M. (2002). Anxiety, arousal and visual attention: A mechanistic account of performance
variability. Journal of Sports Sciences, 20(3), 237251.
https://doi.org/10.1080/026404102317284790
Jetter, M., & Walker, J. K. (2015). Game, set, and match: Do women and men perform differently in
competitive situations? Journal of Economic Behavior & Organization, 119, 96108.
https://doi.org/10.1016/j.jebo.2015.07.017
Latinjak, A. T., Torregrossa, M., Comoutos, N., Hernando-Gimeno, C., & Ramis, Y. (2019). Goal-
directed self-talk used to self-regulate in male basketball competitions. Journal of Sports
Sciences, 37(12), 14291433. https://doi.org/10.1080/02640414.2018.1561967
Accepted for publication in Psychology of Sport and Exercise 25/05/21 PREPRINT
19
Leary, M. R., Terry, M. L., Batts Allen, A., & Tate, E. B. (2009). The Concept of Ego Threat in Social
and Personality Psychology: Is Ego Threat a Viable Scientific Construct? Personality and
Social Psychology Review, 13(3), 151164. https://doi.org/10.1177/1088868309342595
Liu, J., Shen, K., & Li, H. (2019). How state anxiety and attentional bias interact with each other: The
moderating effect of cognitive appraisal. Attention, Perception, & Psychophysics, 81(3), 694
706. https://doi.org/10.3758/s13414-018-01650-y
Marozzi, M., Mukherjee, A., & Kalina, J. (2020). Interpoint distance tests for high-dimensional
comparison studies. Journal of Applied Statistics, 47(4), 653665.
https://doi.org/10.1080/02664763.2019.1649374
Martens, R., Vealey, R. S., & Burton, D. (1990). Competitive Anxiety in Sport. Human Kinetics.
Masaki, H., Maruo, Y., Meyer, A., & Hajcak, G. (2017). Neural Correlates of Choking Under Pressure:
Athletes High in Sports Anxiety Monitor Errors More When Performance Is Being Evaluated.
Developmental Neuropsychology, 42(2), 104112.
https://doi.org/10.1080/87565641.2016.1274314
Masters, R., & Maxwell, J. (2008). The theory of reinvestment. International Review of Sport and
Exercise Psychology, 1(2), 160183. https://doi.org/10.1080/17509840802287218
Masters, R. S. W. (1992). Knowledge, knerves and know-how: The role of explicit versus implicit
knowledge in the breakdown of a complex motor skill under pressure. British Journal of
Psychology, 83(3), 343358. https://doi.org/10.1111/j.2044-8295.1992.tb02446.x
Mesagno, C., & Mullane-Grant, T. (2010). A Comparison of Different Pre-Performance Routines as
Possible Choking Interventions. Journal of Applied Sport Psychology, 22(3), 343360.
https://doi.org/10.1080/10413200.2010.491780
Metcalfe, J. (2017). Learning from Errors. Annual Review of Psychology, 68(1), 465489.
https://doi.org/10.1146/annurev-psych-010416-044022
Moser, J., Moran, T., Schroder, H., Donnellan, B., & Yeung, N. (2013). On the relationship between
anxiety and error monitoring: A meta-analysis and conceptual framework. Frontiers in Human
Neuroscience, 7. https://doi.org/10.3389/fnhum.2013.00466
Nicholls, A. R., Holt, N. L., Polman, R. C. J., & James, D. W. G. (2005). Stress and Coping Among
International Adolescent Golfers. Journal of Applied Sport Psychology, 17(4), 333340.
https://doi.org/10.1080/10413200500313644
Nieuwenhuys, A., & Oudejans, R. R. D. (2012). Anxiety and perceptual-motor performance: Toward an
integrated model of concepts, mechanisms, and processes. Psychological Research, 76(6), 747
759. https://doi.org/10.1007/s00426-011-0384-x
Otten, M. (2009). Choking vs. Clutch Performance: A Study of Sport Performance under Pressure.
Journal of Sport and Exercise Psychology, 31(5), 583601.
https://doi.org/10.1123/jsep.31.5.583
Payne, K. L., Wilson, M. R., & Vine, S. J. (2018). A systematic review of the anxiety-attention
relationship in far-aiming skills. International Review of Sport and Exercise Psychology, 0(0),
1–31. https://doi.org/10.1080/1750984X.2018.1499796
Pocock, C., Bezodis, N. E., Davids, K., & North, J. S. (2018). Hot hands, cold feet? Investigating
effects of interacting constraints on place kicking performance at the 2015 Rugby Union World
Cup. European Journal of Sport Science, 18(10), 13091316.
https://doi.org/10.1080/17461391.2018.1486459
Roberts, L. J., Jackson, M. S., & Grundy, I. H. (2017). Choking under pressure: Illuminating the role of
distraction and self-focus. International Review of Sport and Exercise Psychology, 0(0), 121.
https://doi.org/10.1080/1750984X.2017.1374432
Accepted for publication in Psychology of Sport and Exercise 25/05/21 PREPRINT
20
Schüler, J., & Langens, T. A. (2007). Psychological Crisis in a Marathon and the Buffering Effects of
Self-Verbalizations1. Journal of Applied Social Psychology, 37(10), 23192344.
https://doi.org/10.1111/j.1559-1816.2007.00260.x
Schweickle, M. J., Vella, S. A., & Swann, C. (2021). Exploring the “clutch” in clutch performance: A
qualitative investigation of the experience of pressure in successful performance. Psychology of
Sport and Exercise, 54, 101889. https://doi.org/10.1016/j.psychsport.2021.101889
Solomonov, Y., Avugos, S., & Bar-Eli, M. (2015). Do clutch players win the game? Testing the validity
of the clutch player’s reputation in basketball. Psychology of Sport and Exercise, 16, 130138.
https://doi.org/10.1016/j.psychsport.2014.10.004
Swann, C., Crust, L., Jackman, P., Vella, S. A., Allen, M. S., & Keegan, R. (2017). Performing under
pressure: Exploring the psychological state underlying clutch performance in sport. Journal of
Sports Sciences, 35(23), 22722280. https://doi.org/10.1080/02640414.2016.1265661
Theodorakis, Y., Hatzigeorgiadis, A., & Chroni, S. (2008). Self-Talk: It Works, but How? Development
and Preliminary Validation of the Functions of Self-Talk Questionnaire. Measurement in
Physical Education and Exercise Science, 12(1), 1030.
https://doi.org/10.1080/10913670701715158
Thomas, O., Lane, A., & Kingston, K. (2011). Defining and Contextualizing Robust Sport-Confidence.
Journal of Applied Sport Psychology, 23(2), 189208.
https://doi.org/10.1080/10413200.2011.559519
Toma, M. (2017). Missed Shots at the Free-Throw Line: Analyzing the Determinants of Choking Under
Pressure. Journal of Sports Economics, 18(6), 539559.
https://doi.org/10.1177/1527002515593779
Williams, A. M., & Wigmore, T. (2020). The Best: How Elite Athletes Are Made. Nicholas Brealey
Publishing.
Wilson, M. (2008). From processing efficiency to attentional control: A mechanistic account of the
anxietyperformance relationship. International Review of Sport and Exercise Psychology,
1(2), 184201. https://doi.org/10.1080/17509840802400787
... Basketball [46,[57][58][59] Tennis [60] Volleyball [17,18,55,[61][62][63][64][65] Score differences Closed matches Scoring Rates Balance/unbalance perception Basketball [25,28,31,32,[66][67][68] Football [35,[69][70][71] Tennis [72][73][74] Volleyball [4,11,56,[61][62][63][64][65][75][76][77] Scoring rhythm Score-line Trend Match Status Game Status Current on the scoreboard Basketball [59,78,79] Football [14,70,71,80,81] Hockey [38,82] Several Sports [83][84][85] Tennis [86] Volleyball [11,61,62,87,88] Previous period results Previous set result Previous action result Basketball [89] Football [14] Tennis [19,60,74,[90][91][92][93] Several Sports [83,94] Volleyball [64,87,88] Competitive Load Basketball [27,95] Several Sports [20,43,[96][97][98] Tennis [19,60,99] Volleyball [11,62,63,76,100,101] Scoring First Football [50,102] Hockey [38] Tennis [19,92,99] Match Congestion Basketball [32] Tennis [103] Type of a match ...
Full-text available
Article
The main objective was to know the elite coaches’ opinions regarding the relevance, definition, and importance of volleyball contextual variables to measure individual performance in competition. After performing a literature review, an instrument to gather the opinion of the world’s elite volleyball coaches was elaborated by four volleyball specialists. The sample of experts consisted in 20 world’s elite volleyball coaches who met at least three experience years in first division or national teams. The instrument collected experts’ information on the contextual variables in relation to relevance, definition, and importance. Cronbach’s α and Aiken’s V coefficient were used to test the reliability and content validity of the contextual variables, respectively. To compare the importance of the contextual variables U de Mann-Whitney and Kruskal-Wallis tests were used. Results showed that opposition level, set period, score difference, results of the previous set, competitive load variables, high level, final periods ≥20, and high load categories were relevant (Aikens V > 0.70). In addition, high level, final period ≥20 and ≥+10, and high load categories were significantly more important (p < 0.05). We conclude that, according to the elite coaches, the contextual variables should not be analyzed separately. Future studies should consider contextual variables dynamically.
... In this regard, the ability of the players to deal with pressure situations is directly related to their performance [3,4]. Furthermore, the probability of an error in performance increases significantly with the pressure of the point [5]. Thus, previous studies have explored anxiety in tennis among different age groups, playing standards, as well as gender differences, using a variety of assessment tools [6][7][8]. ...
Full-text available
Article
To study the influence of age and gender on pre-competitive and post-competitive anxiety and self-confidence in young tennis players. A total of 42 U'12 to U'18 category tennis players, 12 females and 30 males, participated in this cross-sectional study. The players had a mean age of 13.74 (2.07) years old and a national competitive experience of 4.00 (2.14) years. The pre-competitive anxiety of the participants was assessed using the Competitive State Anxiety Inventory-2R (CSAI-2R) and the State Trait Anxiety Inventory (STAI-E) before and after an official tournament's match. Results showed that younger players showed lower trait anxiety (r = 0.333; p < 0.05), lower pre-match state anxiety (r = 0.501; p< 0.01) and lower pre-match somatic anxiety (r = 0.313; p < 0.05). Furthermore , girls exhibited higher values of state anxiety (r = 0.445; p < 0.01) and somatic anxiety (r = 0.440; p < 0.01) than boys before the match. However, differences were not observed in the trait anxiety measured by STAI-E (r = 0.203; p = 0.213), cognitive anxiety (r = 0.140; p = 0.363), and self-confidence measured by the CSAI questionnaire (r = 0.150; p = 0.333), before the match. Therefore, coaches and sport psychologists should implement adequate on-and off-court individualized interventions to manage anxiety, specifically in girls and players over 14 years old. Although anxiety levels were similar to those before the COVID-19 pandemic, due to the influence of the pandemic on mental health, results might be taken with caution.
... Anxiety among the players playing at different positions either as guards, forward or centre players was observed to be not influencing their pre-competition anxiety because every team members are dependent on each other for support. This is different in comparison to the pre-competition anxiety be it for the subscale of cognitive or somatic anxiety among athletes who play individually such as in the field or racetrack (Nandu & Noordin, 2020) and tennis (Harris et al., 2021). ...
Full-text available
Article
Psychological factors are some of the factors influencing a player's achievement in basketball tournaments. A survey study was conducted to compare the pre-competition anxiety level based on the players’ positioning among 144 male basketball players who are under 18 years old in Sabah. The Competitive State Anxiety Inventory-2 (CSAI-2) which was translated into the Malay language was used as the main instrument in this study along with the Kruskal-Wallis Test. The findings indicated that there is no significant difference in the pre-competition anxiety level among the players based on their positioning χ2(2) = 3.742, p = .154 (p> .05). The findings of the cognitive anxiety subscale also were found to be insignificant based on the positioning; χ2(2) = 1.938, p = .380 (p> .05). The somatic anxiety of the pre-competition findings also showed no difference based on the positions of players; χ2(2) = 1.509, p = .470 (p> .05). Moreover, the values for subscale of confidence was also found to be insignificant; χ2(2) = 1.565, p = .457 (p> .05). Therefore, the study found out that the environmental factors play a crucial role in reducing the basketball players’ pre-competition anxiety. The coaches and trainers should consider these environmental factors in improving their players’ pre-competition anxiety, thus allowing them to compete and perform at their best state of mind.
Full-text available
Article
Quiet Eye (QE) consists of an ocular fixation (at least 100 ms) that precedes specific athletic movements, supporting their accuracy. Evidence extensively supports the role of QE in performance in various sports (e.g., golf, shooting) and shows that a main function of QE is to facilitate the processing of information. In interceptive timing sports, where athletes have to perform two sub-tasks (i.e., accurately intercept a moving object and to control it in order to be able to hit it on a specific target) for a successful shot, longer QE can facilitate the processing of the incoming information. In a within-subjects experimental design, 12 elite table tennis players (mage = 14.66 years, mexperience = 7.33 years) underwent an exercise in which they had to hit a succession of 100 balls, either (a) on the entire opponent’s half (simple condition) or (b) on two narrow spots on the table (complex condition). Results have shown that hit balls were preceded by longer QE than missed balls (p = .035, η²p = .34). Moreover, in the complex condition athletes extended the duration of QE more for scoring a hit, compared to the simple condition (p = .005, η²p = .52) Additionally, in the complex condition we detected a (strong) positive correlation (r(10) =.74, p < .01) between QE duration and performance. Finally, athletes increased the duration of QE more after a missed ball than after a hit ball (p = .059, η2p = .38). This study shows that QE sustains performance in an interceptive timing sport, by helping athletes to process and respond to complex information, and by helping them to recover after missed shots.
Full-text available
Article
The aim of this study was to monitor the mental activation training during match pressure imageries using a protocol with (MT) and without mental training (wMT) performed in the office and on the tennis court based on the analysis of heart rate, brain waves and subjective ratings in a professional tennis player with high imagery experience. Results showed that both in the office (MTo/wMTo) and on the court (MTc/wMTc) the tennis player’s heart rate increased in the match pressure imagery (I.3-8), being higher in the MTo. It decreased in the pressure imagery using mental tools (I.8-13) in the MT. In the case of brainwaves, beta and gamma waves increased in the match pressure imagery (I.3-8); while beta, gamma, delta and theta waves decreased in the pressure imagery using mental tools (I.8-13), being higher in the office. Entropy decreased in the match pressure imagery (I.3-8), being higher in the MTo. It increased in the pressure imagery using mental tools (I.8-13), being higher in the MTo. Regarding subjective ratings, the tennis player felt the pressure in the match pressure imagery, being higher in MT. In the pressure imagery using mental tools he regulated the activation to feel it at an optimal level (7). In the imagery reality, the olfactory and gustatory dimensions were the most difficult to feel in both imageries. Keywords EEG monitoring, mental activation training, pressure, imagery, EEG, tennis
Full-text available
Article
The aim of this study was to monitor the activation during a neutral situation imagery (NSI) and a pressure situation imagery (PSI), based on the analysis of heart rate, brain waves and subjective ratings in athletes. The sample was made up of sixteen professional tennis players. Imagery protocols consisted of 3 phases; the first and the third involved being focused on their deep breathing (2 min.); the second, in the NSI, first service routine (17 sec.-1 min. 21 sec.) and in the PSI, a match pressure situation (2 min. 10 sec.). Results showed that both NSI and PSI increased heart rate. This increase was higher in the PSI and its highest point was at the maximum pressure moment: interval 5-6. In the case of brain waves, both NSI and PSI caused a decrease in gamma wave activity (intervals 3-8). In the PSI, there was also an increase in gamma waves in interval 5-6, the maximum pressure moment. Entropy was lower in the NSI. In regard to subjective ratings, in the psychological skills there were only significant differences in the PSI between pre- and post-activation at the during moment (pressure). In the imagery reality, olfactory and gustatory dimensions were the most difficult to perceive as real in the NSI and only the olfactory in the PSI.
Full-text available
Article
Background Researchers have demonstrated that elevation in state anxiety leads to elevation in attentional bias favouring the processing of negative information, and that the magnitude of this attentional bias change varies across individuals. However, research has not identified the mechanisms that underpin individual variation in state-anxiety induced attentional bias change. Researchers have also demonstrated that inhibitory control of attention becomes impaired when state anxiety is elevated, and cognitive models propose that impaired inhibitory control of attention may underpin attentional bias to negative information. Thus, the present study investigated whether individual differences in the magnitude of attentional bias elevation elicited by heightened state anxiety is predicted by the degree to which such state anxiety elevation impairs attentional control.Methods Eighty participants completed assessments of attentional bias to negative information and inhibitory control of attention prior to, and following, a procedure designed to elevate state anxiety.ResultsIt was observed that greater elevation in attentional bias to negative information was predicted by lesser decline in inhibitory control of attention as state anxiety increased.Conclusions Findings support proposal of a relationship between attentional control and attentional bias to negative information, though are inconsistent with the proposal that heightened attentional bias to negative information is uniformly underpinned by greater impairment in attentional control. Implications are discussed.
Full-text available
Article
The purpose of this study was to explore whether a personalized self-talk intervention influenced mental toughness, rating of perceived exertion, sense of the urge to slow down, perceived performance and finish times in a series of 800-meter run time trials. While mental toughness has been associated with improved endurance performance, the effect of changing an individual’s momentary self-talk on mental toughness and finish time has not yet been examined. This single-subject, multiple baseline design case study incorporated three participants who each ran a series of 11 − 13 maximum effort 800-meter time trials on the track, separated by a minimum of two days, across ten weeks. Following an initial series of four to six baseline sessions, they were each then provided a personalized self-talk intervention before running the seven additional sessions. Visual analysis (including review of non-overlapping data points between baseline, intervention, and follow-up sessions) demonstrated the personalized self-talk intervention positively influenced mental toughness and finish times across all three participants but did not consistently affect the rating of perceived exertion, urge to slow down or perceived performance. Additional insights were identified through the integration of social validation interviews informally after each run session and then formally after the intervention. These insights included identifying a new baseline of effort accompanied by different levels of mental toughness and an intrigue on the part of participants about the notable improvement in outcomes in spite of previously perceived “all-out” effort. Lay Summary: Mental toughness variability and 800 meter finish times were both positively influenced by a personalized self-talk intervention in runners. In addition, as mental toughness increased, 800 meter finish times improved.
Full-text available
Article
Background and Objectives: While the potentially negative effects of pressure on skilled performance have been well studied in laboratory-based research, theoretically driven questions based on real-world performance data are lacking. Design: We aimed to test the predictions of the newly developed Attentional Control Theory: Sport (ACTS), using archived play-by-play data from the past seven seasons of the National Football League (American Football). Methods: An additive scoring system was developed to characterize the degree of pressure on 212,356 individual offensive plays and a Bayesian regression model was used to test the relationship between performance, pressure and preceding negative outcomes, as outlined in ACTS. Results: There was found to be a clear increase in the incidence of failures on high pressure plays (odds ratio = 1.20), and on plays immediately following a previous play failure (odds ratio = 1.09). Additionally, a combined interactive effect of previous failure and pressure indicated that the feedback effect of negative outcomes was greater when pressure was already high (odds ratio = 1.10), in line with the predictions of ACTS. Conclusions: These findings reveal the importance of exploring momentary changes in pressure in real-world sport settings, and the role of failure feedback in influencing the pressure-performance relationship.
Full-text available
Article
This study examined how goal-directed self-talk may help basketball players to self-regulate in stereotypical competitive situations: seconds before a challenging game, while clearly winning or clearly losing, and at the close of a tight game. Participants were recruited in groups of three to four, until preliminary inspection of the data indicated that data saturation was reached. In the end, 34 basketball players voluntarily took part in individual interviews, writing up to three self-instructions they had used in each of the four competitive situations to self-regulate. Content analyses revealed that self-talk in competitive basketball situations serves cognitive functions (e.g., regulating cognition and behaviour), motivational functions (e.g., promoting mastery goals) and emotion and activation-regulating functions (e.g., creating activated states). More specifically, the results also indicated that athletes’ self-talk may serve functions specific to the psychological demands experienced in each situation. It is argued that knowing how athletes counsel themselves, could prove important for applied sport psychologists to design psychological skill training.
Full-text available
Article
Place kicks in Rugby Union present valuable opportunities to score points outside the spatiotemporal dynamics of open play but are executed under varying performance constraints. We analysed effects of specific task constraints and relevant contextual factors on place kick performance in the 2015 Rugby Union World Cup. Data were collected from television broadcasts for each place kick. In addition to kick outcomes, contextual factors, including time of the kick in the match, score margin at the time of the kick, and outcome of the kicker’s previous kick, were recorded. Effects of spatial task constraints were analysed for each kick, using distance (m) and angle (°) of the kick to the goalposts. A binomial logistic regression model revealed that distance from, and angle to, the goalposts were significant predictors of place kick outcome. Furthermore, the success percentage of kickers who missed their previous kick was 7% lower than those who scored their previous kick. Place kick success percentage in the 10 minutes before half-time was 8% lower than the mean tournament success percentage, which was 75% (95% CI 71–78%). The highest kick success percentage was recorded when scores were level (83%; 95% CI 72–91%). Our data highlighted how subtle changes in task constraints and contextual factors can influence performance outcomes in elite performers in international competition. Fluctuations in place kick success suggested that individual constraints, such as thoughts, emotions and fatigue, induced during competition, could interact with perceptions to influence emergent performance behaviours.
Article
Objectives The term “clutch” is colloquially used to describe important or crucial situations in sport, with clutch performance referring to successful performances during these pressurised circumstances. Positioning the clutch as an objective, situational variable, however, may not account for the athletes' subjective appraisal of such situations. Accordingly, this study aimed to explore athletes' perceptions of clutch situations, and further, how these perceptions influenced their performance. Method Sixteen athletes (Mage = 26.88 years) participated in event-focused, semi-structured interviews soon after a clutch performance (M = 4 days later). Data were analysed utilising reflexive thematic analysis. Results Four themes were generated: (1) the clutch involves situational and subjective factors, suggesting that the appraisal of the clutch is influenced by situational and subjective components; (2) the perception of the clutch comes and goes, suggesting that there may be multiple, fluctuating episodes of the clutch within an event; (3) pressure affects performance, and performance affects pressure, suggesting that the appraisal of pressure was perceived to impact performance, and that performance also influenced appraisal of pressure; and, (4) experience of anxiety during the clutch is varied, suggesting that the experience of anxiety is not inherent to clutch performance. Conclusions The clutch has traditionally been considered an objective, situational variable. This study suggests, however, that the clutch relies upon the athlete appraising pressure in response to these situational variables, which may not always occur. Further, other subjective factors may increase the appraisal of pressure, suggesting that the clutch cannot solely be considered as a situational variable.
Article
Modern data collection techniques allow to analyze a very large number of endpoints. In biomedical research, for example, expressions of thousands of genes are commonly measured only on a small number of subjects. In these situations, traditional methods for comparison studies are not applicable. Moreover, the assumption of normal distribution is often questionable for high-dimensional data, and some variables may be at the same time highly correlated with others. Hypothesis tests based on interpoint distances are very appealing for studies involving the comparison of means, because they do not assume data to come from normally distributed populations and comprise tests that are distribution free, unbiased, consistent, and computationally feasible, even if the number of endpoints is much larger than the number of subjects. New tests based on interpoint distances are proposed for multivariate studies involving simultaneous comparison of means and variability, or the whole distribution shapes. The tests are shown to perform well in terms of power, when the endpoints have complex dependence relations, such as in genomic and metabolomic studies. A practical application to a genetic cardiovascular case-control study is discussed.
Article
In the present study, we conducted four experiments to explore how state anxiety influences attentional bias, and vice versa, as well as the moderating effect of cognitive appraisal in this relationship. Experiment 1 focused on whether induced state anxiety could lead to attentional bias. Experiment 2 explored the influence of attentional bias on state anxiety under stressful conditions. Experiments 3 and 4 investigated the moderating effect of cognitive appraisal on the interaction between state anxiety and attentional bias. Our main findings were that state anxiety directly leads to attentional bias, whereas negative attentional bias increases state anxiety under stressful conditions. Moreover, cognitive appraisal moderates the influence of attentional bias on state anxiety, but not the reverse influence. The implications of our study are that it provides empirical evidence for the interaction between state anxiety and attentional bias, and also that it offers insight into the different moderating effects of cognitive appraisal on the relationship.
Article
Theoretical accounts of the anxiety and motor performance relationship cite disruptions to attention as a critical mediating factor. The aims of this paper were to (1) systematically review published research examining attentional mechanisms underpinning the anxiety–performance relationship in targeting skills, and (2) subsequently discuss these findings in relation to contemporary theoretical perspectives. Adhering to PRISMA guidelines, three electronic databases (PubMed, PsycInfo, and SPORTDiscus) were searched from inception until June 2017. Thirty-four articles satisfied the inclusion criteria. Overall, the research is of high methodological quality; however, there is a tendency to focus on the historical dichotomy between self-focus and distraction accounts, whereas empirical support for more contemporary theoretical perspectives is lacking. Whilst this review provides further support for the role of attentional disruptions in anxiety-induced performance degradation, the exact mechanisms still lack consensus. In addition, more innovative experimental designs and measures are required to progress our understanding of moderating variables.