Effects of a Motivational Self-Talk Intervention for Endurance Athletes Completing an Ultramarathon

Abstract

This study examined the effects of strategic, motivational self-talk for runners completing a 60-mile, overnight ultramarathon using a randomised, controlled experiment. Data were collected before, during, and after an annual ultramarathon. Twenty-nine ultramarathon runners were randomly allocated to a motivational self-talk group or an alternative control group. A condition-by-time mixed ANOVA indicated that learning to use motivational self-talk did not affect pre-event self-efficacy or perceived control. A t-test and magnitude-based inference indicated that motivational self-talk did not affect performance. Nevertheless, follow-up data suggested that most participants found the intervention helpful and continued to use it six months after their research commitment, particularly in endurance events and to a lesser extent in training. Participants continued to use self-talk to cope with exertion, as well as other stressors such as blister discomfort and adverse conditions. Suggestions are offered for future research examining the effects of psychological interventions on performance in endurance events.

Full text

1
Abstract
This study examined the effects of strategic, motivational self-talk for runners completing a
60-mile, overnight ultramarathon using a randomised, controlled experiment. Data were
collected before, during, and after an annual ultramarathon. Twenty-nine ultramarathon
runners were randomly allocated to a motivational self-talk group or an alternative control
group. A condition-by-time mixed ANOVA indicated that learning to use motivational self-
talk did not affect pre-event self-efficacy or perceived control. A t-test and magnitude-based
inference indicated that motivational self-talk did not affect performance. Nevertheless,
follow-up data suggested that most participants found the intervention helpful and continued
to use it six months after their research commitment, particularly in endurance events and to a
lesser extent in training. Participants continued to use self-talk to cope with exertion, as well
as other stressors such as blister discomfort and adverse conditions. Suggestions are offered
for future research examining the effects of psychological interventions on performance in
endurance events.

2
Effects of a Motivational Self-Talk Intervention for Endurance Athletes
Completing an Ultramarathon
Psychological skills training (PST) can enhance endurance performance. Although
self-talk, imagery, goal setting, and packages including multiple PST strategies consistently
enhance endurance performance in non-competitive laboratory and field-based endurance
tasks, few studies have examined the effects of psychological interventions such as PST on
performance in actual endurance events, and these studies showed equivocal intervention
effects (McCormick, Meijen, & Marcora, 2015). People typically perform better in endurance
tasks when competing (McCormick et al., 2015). Further, athletes need to manage additional
stressors, and associated emotions, before and during events (e.g., McCormick, Meijen, &
Marcora, 2016; Simpson, Post, Young, & Jensen, 2014), compared to those typically present
in experimental conditions. It is therefore important to demonstrate that the performance
benefits of PST generalise to actual endurance events. To date, few studies of endurance and
other sport performances have measured the effects of a psychological intervention on the
competitive performances of athletes using a randomised, controlled experiment (G. L.
Martin, Vause, & Schwartzman, 2005). Randomised, controlled experiments are the most
rigorous experiments for demonstrating a cause-and-effect relationship between an
intervention and an outcome (e.g., Shadish, Cook, & Campbell, 2002). The present study is
the first randomised, controlled experiment to examine the effects of PST on performance in
an actual endurance event (McCormick et al., 2015). Specifically, this study examined the
effect of strategic, motivational self-talk on performance in an ultramarathon.
Self-talk can be defined as what people say to themselves silently in their head or
aloud, automatically or strategically, to stimulate, direct, react, and evaluate events and
actions (Hatzigeorgiadis, Zourbanos, Latinjak, & Theodorakis, 2014). As captured by this
definition, self-talk refers to self-addressed verbalisations or statements that serve a range of

3
functions. These self-addressed verbalisations may occur automatically without being
-stematic way as a self-
   -    and exercise psychology literature,
there have been two main approaches to researching self-talk (Hatzigeorgiadis, Zourbanos, et
al., 2014). An observational approach has particularly focused on inherent self-talk and
examined the conte  -talk (e.g., Van Raalte, Morrey, Cornelius, & Brewer,
2015) and the antecedent or consequent correlates of self-talk (e.g., Hatzigeorgiadis &
Biddle, 2008), whereas an experimental approach has examined the effects of strategic
instructional or motivational self-talk, particularly on performance (e.g., Hatzigeorgiadis,
Galanis, Zourbanos, & Theodorakis, 2014). The present study aligns with this latter
approach; it examines the effects of a strategic, motivational self-talk intervention on
psychological variables and endurance performance.
Motivational self-talk was chosen as the intervention for three main reasons. First,
strategic, motivational self-talk has been shown to enhance endurance performance in
randomised, controlled experiments in the laboratory (Blanchfield, Hardy, de Morree,
Staiano, & Marcora, 2014; Wallace et al., 2017). Second, self-talk can be fairly
straightforward to learn and apply. In the endurance performance literature, learning to
strategically use motivational self-talk through a 30-60 minute workbook has led to
substantial performance gains (Barwood, Corbett, Wagstaff, McVeigh, & Thelwell, 2015;
Blanchfield et al., 2014; Wallace et al., 2017). It could therefore be a time-efficient
intervention for a population that often has little free time (Simpson et al., 2014). Third,
motivational self-talk is a versatile strategy that could serve multiple functions during an
ultramarathon. Specifically, motivational self-talk can help performers to persevere through
exertion for longer during endurance performances at a constant workload (Blanchfield et al.,
2014; Wallace et al., 2017) and to sustain a faster pace during self-paced performance

4
(Barwood et al., 2015). In addition, motivational self-talk could help ultramarathon runners to
cope with the many stressors experienced during actual endurance events. Ultramarathon
runners must overcome muscle cramping and injuries, gastrointestinal problems, thoughts
about quitting, and adverse weather and conditions, and they need to carefully pace
themselves to meet cut-off times whilst avoiding premature exhaustion (Antonini Philippe,
Rochat, Vauthier, & Hauw, 2016; Holt, Lee, Kim, & Klein, 2014; Simpson et al., 2014). As
the ultramarathon in the present study was completed overnight, additional event-specific
stressors include navigation in the dark and sleep deprivation. Using motivational self-talk as
a coping strategy when experiencing stressors could prevent the stressors from eliciting
negatively-toned emotions and, consequently, from undermining their motivation to sustain
effort and their ability to concentrate on important cues such as navigation cues (e.g.,
Martinent & Ferrand, 2009).
Five broad experiences characterise involvement in ultramarathon running (Simpson
et al., 2014): (a) the support, common personal bonds, and camaraderie of being part of an
ultramarathon community; (b) physical (e.g., training, nutrition, hydration), tactical (e.g.,
preparing equipment and clothing), and mental aspects (e.g., reflecting on previous
experiences) of preparing for an ultramarathon; (c) using mental skills to manage
performance during a race; (d) having opportunities to push perceived capabilities, to
experience nature, and to have spiritual experiences; and (e) the feelings of accomplishment
and euphoria experienced after completing an ultramarathon. Although research has shed
light on the experiences of ultramarathon runners and the coping strategies ultramarathon
runners use (Acevedo, Dzewaltowski, Gill, & Noble, 1992; Antonini Philippe et al., 2016;
Holt et al., 2014; Simpson et al., 2014), no published studies have examined the effects of
PST on performance in an ultramarathon. Further, no studies have examined the effects of

5
PST on performance in long-distance running events of at least half-marathon distance
(McCormick et al., 2015).
In addition to examining the effect of learning and practising motivational self-talk on
ultramarathon performance, this study examined whether motivational self-talk increased pre-
event self-efficacy and perceived control. When a competition is important to an athlete,
when there is uncertainty about the outcome of the competition, and when the competition
will require effort, the athlete may perceive the competition as a challenge or a threat.
Whether an athlete approaches the competition as a challenge or a threat depends on whether
they evaluate their resources as sufficient to cope with the demands of the competition
(Jones, Meijen, McCarthy, & Sheffield, 2009). An athlete is more likely to approach the
competition as a challenge if they possess high self-efficacy, perceive having control over the
situation, and strive towards achieving goals (Jones et al., 2009). This study examined
whether learning motivational self-talk before an ultramarathon could promote a challenge
response to the ultramarathon by teaching the athletes a skill, or a resource, that they could
use to cope with event demands (e.g., physical and environmental demands, encountered
problems) and finish the event (i.e., potentially enhancing their self-efficacy), and by helping
the athletes to perceive themselves as having sufficient control over their responses to the
event demands (i.e., potentially enhancing their perceived control).
The present study examined the effects of a strategic, motivational self-talk
intervention for endurance athletes completing a 60-mile, overnight ultramarathon. There
were four aims. The first aim was to examine whether learning and practising motivational
self-talk improved performance in an ultramarathon. The second aim was to examine whether
learning and practising self-talk increased pre-event self-efficacy and perceived control. The
third aim was to determine whether participants were still using self-talk six months after
receiving the intervention, as well as the functions of this self-talk, which could indicate

6
whether they found learning self-talk valuable. Studies examining the effects of PST on
endurance performance have not reported whether participants continued to use the taught
intervention after they finished their commitment to the research (McCormick et al., 2015).
The final aim was to highlight psychological aspects of participating in an ultramarathon,
such as pre-event and post-event emotions, encountered stressors, and the intensities of
perceived effort, exercise-induced muscle pain, and injury pain experienced over the course
of an ultramarathon. As well as helping the reader to understand the nature of this particular
ultramarathon, such data could inform future interventions that aim to help ultramarathon
runners to cope with the demands of their sport (Simons, 2012).
Methods
Design
A randomised, controlled, posttest-only experimental design was used for the
performance dependent variable.1 For self-efficacy, perceived control, and expectations of
performance improvement, pre- and post-intervention comparisons were made. After
matching for estimated aerobic fitness, participants were randomly assigned to an
experimental group (motivational self-talk) or an alternative control treatment group
(concentration grid) by tossing a coin. Alternative control treatments are similar in duration,
perceived value, and procedure to the experimental treatment, but they target different
outcomes (W. Borg, 1984).
The Ultramarathon
The ultramarathon is an annual 60-mile race along the Thames Path in South England.
Participants navigate a flat, rural terrain that passes through villages. The ultramarathon is
self-supported; participants carry food, water, and clothing, and they navigate using head
torches and supplied route maps. Water and medical attention are available at three
checkpoints that are spaced 15 miles apart. Data were collected before, during, and after three

7
ultramarathons (May 2014, May 2015, and June 2016). The ultramarathons began at 8pm,
and there were cut-off times for each checkpoint.
Participants
All 148 runners who registered for the ultramarathon on any of the three years were
invited to participate. Thirty-two runners (22%) volunteered to participate in the study.
Fifteen were randomly assigned to the self-talk group, and 17 were assigned the control
group. Fourteen participants in the self-talk group attended the event and were included in the
analyses, and 15 participants in the control group attended the event. Three participants
dropped out of the study because they decided not to run the ultramarathon (n = 2) or because
they had insufficient time for the intervention workbook and did not want to complete rating
scales during the ultramarathon (n = 1) (a flowchart is available on request). Twenty-five of
the 29 participants were male, and 25 were British (M age = 39.3 years, SD = 8.4; M height =
177 cm, SD = 10; M weight = 73.5 kg, SD = 8.1; M weekly training hours = 8.9, SD = 3.6; M
events of at least half-marathon distance during the previous 12 months = 5.7, SD = 3.8; M
 
2max = 50.9 ml·kg-1·min-1, SD = 8.3; M  
2max = 43.0
ml·kg-1·min-1, SD = 4.1). The four non-British participants were fluent in the English
language. Achievement goals, in relation to an important endurance event, were measured in
Survey 1 using the Achievement Goal Questionnaire for Sport (AGQ-S) (Conroy, Elliot, &
Hofer, 2003). Scores on the mastery-approach goals (M = 17.0, SD = 2.99) subscale were
greater than scores on the mastery-avoidance (M = 11.3, SD = 5.57), outcome-approach (M =
10.4, SD = 4.17), and outcome-avoidance subscales (M = 10.3, SD = 5.12) (scores can range
from 3 to 21).
Participants had little prior experience with PST, particularly using self-talk
systematically. During an intake interview, no participants reported past experience working
with a sport psychologist. When describing the mental strategies they use during endurance

8
events, six participants referred to their self-talk statements (two were assigned to the self-
talk group and four to the control group), including motivational content. The most common
psychological strategy (referred to by 15 participants) was race chunking, which involves
breaking the distance down into shorter, more manageable distances.
Procedure
Ethical approval was granted by the department ethics committee, and informed
consent was obtained. Figure 1 provides an overview of the timeline, including data
collection dates and intervention delivery dates. Self-report data were mainly collected using
four online surveys distributed by email. One battery of questionnaires was completed on
paper 30-90 minutes before the ultramarathon start. With consideration to missing self-report
data, all participants completed Survey 1, 26 participants (90%) completed Survey 2 (two
participants in the self-talk group and one in the control group did not), 26 participants (90%)
completed all of the pre-event questionnaires (three participants in the self-talk group arrived
late), all participants completed Survey 3, and 27 participants (93%) completed Survey 4 (one
participant in each group did not). Twenty-three participants (79%) completed all of the
surveys. Mean values were calculated using existent data.
[Insert Figure 1]
Variables
 
2max 
2max was estimated using data collected in Survey 1 and
corresponding mathematical formulas for males (Malek, Housh, Berger, Coburn, & Beck,
2005) and females (Malek, Housh, Berger, Coburn, & Beck, 2004).
Self-efficacy and perceived control. A self-efficacy scale was designed following
the guidance of Bandura (2006). Participants rated how certain they were (0-100 degree of
confidence) that they could do each of 10 things during the ultramarathon, which reflected
anticipated demands (e.g.,        ,

9
). MHighly
               
Participants also rated how much control they perceived their self to have over each of the 10
NC
self-efficacy and perceived control values were mean values of the 10 ratings. These scales
were completed in Survey 1 and at the ultramarathon before the start.
Performance expectations. Expectations of performance improvement were
measured pre-randomisation (Survey 1) and before the ultramarathon start to determine the
possibility of a placebo effect. Participants rated their degree of agreement with a statement
about the potential performance-enhancing effects of psychological interventions on a seven-

Pre-event motivation. Before the ultramarathon start, participants rated their

              -point
 
Sport emotions. Participants completed the Sport Emotion Questionnaire (Jones,
Lane, Bray, Uphill, & Catlin, 2005) before the ultramarathon start, and they retrospectively
rated how they felt after they finished the ultramarathon (Survey 4). This questionnaire
assesses anger, anxiety, dejection, excitement, and happiness. The fatigue items of the Brunel
Mood Scale (Terry, Lane, Lane, & Keohane, 1999) were included among items. Both
measures use the same five-      N    

Perception of effort and pain. Nine research points were marked on 
route maps, near to identifiable landmarks. At each, participants were asked to rate their
perception of effort, exercise-induced pain, and injury-related pain by marking a value onto a

10
scale. Participants rated their overall perception of effort using the 15-point rating of
perceived exertion (RPE) scale (G. A. Borg, 1998), and they rated the intensity of exercise-
induced pain and injury-related pain using 0-10 pain scales   
Smith, & Lee, 1997). Perception of effort was defined as how effortful, heavy, and strenuous
the exercise feels. Exercised-induced pain was defined as the pain that is produced by muscle
burn and ache as a result of repeated or prolonged muscular contraction, whereas injury-
related pain was defined as the pain that may result from injury (e.g. blisters, twisted ankle).
Sixteen participants (55%) recorded all RPE and pain values applicable to them, and an
additional four participants (14%) recorded most RPE and pain values.
Performance. 
minute. Performance times were downloaded from the official website of the ultramarathon.
Manipulation checks. Following the ultramarathon (Survey 3), questions determined
whether participants in the self-talk group used the workbook to identify self-talk statements
(Yes/No), to what extent these participants used self-talk statements during the ultramarathon
(five-point Likert scale), whether self-talk statements were helpful (Yes/No), and the
occasions and purposes of using self-talk statements (qualitative data). Participants in the
concentration-grid group used the same five-     N   A great
-talk statements during the event.
Contamination checks. Following the ultramarathon (Survey 3), contamination
questions were used to determine whether participants were aware of the other intervention.
Social validity. Social validation is used to determine satisfaction with an
intervention (Page & Thelwell, 2013). Following the ultramarathon (Survey 3), participants
used a five-point Likert scale to rate the extent to which an improvement in long-distance
running performance is important to them, the extent to which an improvement in
concentration is important to them (control group only), the extent to which the intervention-

11
delivery methods were acceptable, the extent to which they viewed changes in their long-
distance running performance to be significant, and the extent to which they viewed changes
           S   
S      ultramarathon (Survey 4), participants re-rated the
extent to which they viewed changes in their performance and concentration to be significant.
Follow-up of intervention use. Six months after the event, participants reported
whether they were still using self-talk statements (self-talk group) or concentration grids
(control group) using a five-            
Participants in the self-talk group were asked whether they used self-talk statements in
training during the previous month and during their most recent long-distance running event.
They also provided information on the occasions and purposes of using self-talk statements.
Interventions
Intake interview. An intake interview was conducted with each participant to build
rapport before distributing the intervention workbooks and to gain information relevant to the
study. The primary researcher ran 20 intake interviews by video call and nine by telephone.
Using a set protocol, the researcher and participant discussed the following topics: the
         
expectations and goals for the ultramarathon; what sport psychology is and the parti
experiences with it; psychological strategies already used by participants; and the format of
the workbook. Participants were asked not to discuss the content of their workbook with
other competitors, including competitors in later years. The mean intake duration was 39
minutes (SD = 16).
Workbooks. To the extent possible, the workbooks were presented in a similar
format. Both workbooks included educational material, exercises for participants to complete,

12
and four practice logs that would each take up to five minutes to complete. The workbooks
were distributed one day after the final intake interview.
Self-talk workbook. Learning to strategically use motivational self-talk through a 30-
minute workbook, followed by two weeks of practising self-talk during exercise bouts and
refining self-talk statements, has been shown to benefit endurance performance (Blanchfield
et al., 2014). Following a brief introduction to self-talk statements, participants were asked to
notice their self-talk during a training run, and they recorded the effects that self-talk
statements had on how they felt. Participants then compared their statements to 32
motivational statements that were located in the self-talk literature (Blanchfield et al., 2014;
Miller & Donohue, 2003) or PST consultancy notes, and they picked four statements from the
two lists that would be valuable during the beginning, middle, or later stages of the
ultramarathon (Blanchfield et al., 2014). Participants were encouraged to use motivational
statements to counter thoughts about withdrawing effort     
   
encouraged to practise and refine statements in training until the ultramarathon. Participants
were asked to complete four logs detailing their use of self-talk (Blanchfield et al., 2014).
Concentration workbook. The importance of concentration in sport was briefly
introduced. The workbook then introduced the concentration grid and explained how
participants could use it to develop their concentration. Participants were encouraged to
complete at least two concentration grids each day, on as many days as possible, until the
ultramarathon, and they were asked to practise in both quiet and distracting environments.
Participants were given 20 concentration grids, and they were asked to complete four logs
detailing their use of the concentration grid.
Compliance checks. Participants completed a survey before the ultramarathon
(Survey 2) to determine whether they had read the workbook and were practising their

13
allocated strategy. Participants were given the opportunity to ask questions. Answers to
questions were sent by email to all participants using the same workbook.
Data Analysis
A one-tailed, independent-samples t-test was used to determine if there was a
statistically-significant difference in the performance times between groups. To address the
practical significance of the effect on performanceparticularly considering the small
sample size and posttest-only designthe probabilities that the true effect size is beneficial
( d > 0.20), trivial (between ± 0.20), or harmful (< -0.20) were calculated using a
magnitude-based inferences spreadsheet (http://sportsci.org/resource/stats/generalize.html)
(Batterham & Hopkins, 2006). Condition-by-time (2x2) mixed ANOVAs were used to
determine whether the intervention influenced changes in self-efficacy and perceived control.
Partial eta squared (p2) effect sizes are presented for these ANOVAs (small, moderate, and
large effect size anchors are 0.01, 0.06, and 0.14, respectively). Mann-Whitney U tests were
used to compare pre-post changes in expectations of performance improvement and pre-event
motivation values between groups. Qualitative data from Survey 3 and Survey 4 were
organised into themes.
Results
Self-Efficacy
For participants in the self-talk group, self-efficacy decreased from a mean of 88.7
(SD = 7.44) when Survey 1 was completed (pre-randomisation) to 83.5 (SD = 9.85) (out of
100) at the ultramarathon before the start. For participants in the control group, self-efficacy
increased from 86.9 (SD = 10.2) to 87.5 (SD = 9.41). The main effect of time, F(1, 24) =
1.74, p = .20, p2 = 0.068, the main effect of condition, F(1, 24) = 0.12, p = .73, p2 = 0.005,
and the interaction between time and condition, F(1, 24) = 2.70, p = .11, p2 = 0.10, were not
statistically significant.

14
Perceived Control
For participants in the self-talk group, perceived control decreased from a mean of
8.46 (SD = 0.95) when Survey 1 was completed to 7.89 (SD = 1.62) (out of 10) at the event
before the start. For participants in the control group, perceived control decreased from 8.73
(SD = 1.31) to 8.33 (SD = 1.12). The main effect of time was statistically significant, F(1, 24)
= 5.34, p = .030, p2 = 0.18. The main effect of condition, F(1, 24) = 0.60, p = .45, p2 =
0.024, and the interaction between time and condition, F(1, 24) = 0.15, p = .70, p2 = 0.006,
were not statistically significant.
Performance Expectations
Expectations of performance improvement immediately before the ultramarathon
were slightly higher in the self-talk group (out of 1 to 7, median [Mdn] = 5.5, interquartile
range [IQR] = 5 to 6) than the control group (Mdn  IQR = 4 to 5). Changes in
expectations of performance improvement, from pre- to post-intervention, did not differ
significantly between groups (both Mdns = 0, self-talk IQR = -1 to 0.5, control IQR = -1 to 0),
U = 60.0, p = .15.
Pre-Event Motivation
Motivation to participate did not differ significantly between groups (out of 0 to 4,
both Mdns  self-talk IQR = 3 to 3.5, control IQR = 3 to 4), U = 77.5, p =
.55. Motivation to race against others also did not differ significantly (self-talk Mdn  
-talk IQR = 1 to 2.5, control Mdn  IQR = 1 to 2), U
= 74.5, p = 0.46.
Sport Emotions
Emotions and fatigue data are presented in Table 1. Participants generally felt happy
and excited before the ultramarathon and after finishing it, with some pre-start anxiety and
post-event fatigue.

15
[Insert Table 1 here]
Ultramarathon Characteristics
Ten participants (34%) ran with other runners throughout the ultramarathon. The 19
participants who performed alone estimated that they ran alone for a mean of 49% (SD = 36)
of the ultramarathon (i.e., time alone varied greatly between participants). As preparation for
a 184-mile ultramarathon along the route, 11 participants (38%) (six self-talk, five control)
carried some additional equipment with them, and three participants (10%) (one self-talk, two
control) carried all additional equipment with them (e.g., sleeping and cooking equipment).
Mean estimated equipment weights for these participants were 5.6 kg (SD = 1.1) (four self-
talk participants provided estimates of M = 5.3 kg, SD = 1.3, and six control participants
provided estimates of M = 5.8 kg, SD = 1.1). As displayed in Figure 2, RPE, exercise-induced
pain, and injury-related pain increased with distance covered during the ultramarathon.2
Twenty-two of 29 (76%) of all entrants (83% of finishers) experienced new or pre-existing
injuries, including blisters, during the ultramarathon. Excluding injury, 24 of 29 participants
(83%) (11 of 14 self-talk, 13 of 15 control) encountered problems during the ultramarathon.
By far, the most common problem, reported by 16 participants (55%) (six of 14 self-talk, 10
of 15 control), related to difficulty navigating in the dark and associated navigation errors.
[Insert Figure 2 here]
Performance
Eleven of 14 participants in the self-talk group finished the ultramarathon, and 13 of
15 participants in the control group finished. A statistician, who viewed performance times
but was blinded to allocations, suggested only including participants who met the advertised
time limit of 15 hours (20 of 24 finishers) in the analyses. The mean performance time of the
self-talk group (824 minutes, SD = 97) was 12 minutes (1.44%) faster than the control group
(836 minutes, SD = 75), t(18) = 0.31, p = .76. d = 0.13).

16
The probability that the true effect is practically harmful/trivial/beneficial is 22.1/34.6/43.3%.
For transparency, analyses with all finishers included are also presented. The mean time of
the control group (848 minutes, SD = 75) was five minutes (0.57%) faster than the self-talk
group (853 minutes, SD = 114), t(22) = 0.13, p = .90. The effect size was trivial (d =
-0.05). The probability that the true effect is practically harmful/trivial/beneficial is
35.6/37.4/27.0%. The efficacy of self-talk was unclear in both analyses.
Intervention Compliance
Twelve of 14 participants in the self-talk group reported that they used the workbook
to identify self-talk statements, and 10 participants submitted completed workbook logs.
When participants used a five-point (1-5) Likert scale to indicate if they used self-talk during
the ultramarathon, the median response for participants in the self-talk group   
 IQR = 3 to 5, no participants reported no self-talk use), and the median
response for participants in the control      IQR = 1 to 3, eight
participants reported no self-talk use). Twelve of 15 participants in the control group reported
that they had completed concentration grids, and 10 participants submitted completed
workbook logs.
Intervention Contamination
No participants reported being told about the content of the other workbook.
Social Validity
Improvements in long-distance running performance were important to both groups
(out of 1 to 5, self-talk Mdn  IQR = 4 to 5, control Mdn  
IQR = 4 to 5), improvements in concentration were important to the control group (Mdn = 4,
IQR = 4 to 5), and the methods used to deliver the interventions were acceptable to
participants in both groups (both Mdns = 4, self-talk IQR = 4 to 5, control IQR = 3 to 5).
After the ultramarathon, changes in performance were not considered to be significant by

17
either group (both Mdns    , self-talk IQR = 2.5 to 4, control
IQR = 3 to 3), and changes in concentration were not considered to be significant by the
control group (Mdn = 3, IQR = 3 to 4). Eleven of 14 participants in the self-talk group
reported that using self-talk statements was helpful during the ultramarathon. Six months
after the ultramarathon, changes in running performance were considered significant by the
self-talk group (Mdn = 4, IQR = 3 to 4). More participants (8 of 13 respondents) perceived
that self-talk had benefited their performance six months after the ultramarathon, compared
with soon after the ultramarathon (5 of 14 respondents). Changes in performance and
concentration were not considered significant by the control group (Mdns = 3, IQRs = 1 to 4).
Follow-Up of Intervention Use
Six months after the ultramarathon, 11 of 13 participants in the self-talk group
reported intentionally using self-talk in training during the previous month (out of 1 to 5, Mdn
 IQR = 2 to 3). Ten of 11 participants who had participated in a long-
distance running event since the ultramarathon reported intentionally using self-talk during
their most recent event (Mdn 4  IQR = 3 to 4), which was a mean
of 21 weeks (SD = 7) after the ultramarathon. Three of 11 participants in the control group
reported using the concentration grid during the previous month.
Qualitative Data
Participants in the self-talk group reported finding self-talk particularly helpful during
the ultramarathon for different reasons. Reasons reported were: to focus (e.g., on the goal of
completion) (reported by four); to cope with feelings of tiredness (reported by three), pain
(reported by one), and injury (reported by one); and to cope with other stressors, namely
getting lost or going off course (reported by two), low moments (reported by one), and
adverse weather (reported by one). Six months after the ultramarathon, the most common use
of self-talk in training (reported by nine) was to persevere or to push harder, particularly

18
despite tiredness (specifically referred to by six)    -talk] as I felt tired and
unwilling to run, or to make me do a hard effort fasterhe most common uses of self-talk
in events were to persevere or to push harder (reported by five) and to cope with tough
periods of the race (reported by four) that included the following stressors: blisters (reported
by two); navigation errors (reported by one); low moments (reported by one); feeling wet and
cold (reported by one); and adverse weather (reported by one). Comments of seven
participants in the self-talk group indicated that that they had, to some extent, been using self-
talk statements naturally or unsystematically before commencing the study.
Discussion
This research examined the effects of a strategic, motivational self-talk intervention
for endurance athletes completing a 60-mile, overnight ultramarathon. Laboratory research
demonstrates that motivational self-talk can improve endurance performance, but no
randomised, controlled experiments have examined the effects of self-talk (or other PST
interventions) at actual events. The results did not demonstrate a performance benefit. The
sample size was small, however, and performance times varied considerably because of
variations in competitive standard, injury, and navigation errors. These factors reduce
statistical power compared to well-controlled laboratory settings and shorter-duration
endurance tasks. Participants also spent much of the race running with others, meaning that
they may not have chosen their own pace.
Although a performance benefit was not demonstrated, the data suggested that most
participants found the intervention helpful and continued to use it six months after their
commitment to the research, particularly in endurance events and to a lesser extent in
training. Participants continued to use self-talk to cope with exertion, as well as other
stressors such as blister discomfort and adverse conditions. Further, an additional three
participants perceived that self-talk had benefited their performance six months after the

19
ultramarathon, compared with soon after the ultramarathon. These findings highlight the
potential value of self-talk to endurance athletes, because the study did not require the
athletes to continue using self-talk.
Many endurance athletes naturally talk to themselves during endurance events. For
example, Van Raalte, Morrey, Cornelius, and Brewer (2015) found that 88% of marathon
runners engaged in various types of self-talk that included motivational, associative,
dissociative, goal-related, incentive, mantra, and spiritual self-talk. In the present study, the
intake interviews and the data of the control group suggested that, without receiving the self-
talk workbook, participants were mostly unaware of using self-talk. During the intake
interview, only six of 29 participants referred to using self-talk statements as a psychological
strategy during long-distance running events, and eight of 14 participants in the control group
reported using no self-talk (of any type) during the ultramarathon. The self-talk group used
self-talk more often during the ultramarathon than the control group and, after using the
workbook, seven participants in the self-talk group commented that they previously used self-
talk naturally or unsystematically. It is proposed that the workbook used in this study helped
athletes to become aware of their self-talk and its consequences, and encouraged them to use
self-talk more often, and also more systematically.
This study also examined whether learning and practising self-talk increased pre-
event self-efficacy and perceived control. Self-efficacy, perceptions of control, and striving
towards achieving goals influence challenge responses to upcoming competition (Jones et al.,
2009). Participants in the present study strived to achieve mastery goals, which means that
they typically strived to achieve absolute or intrapersonal competence (Conroy et al., 2003).
In other words, participants strived to perform well relative to their own standards, rather than
relative to others. This finding is consistent with research that demonstrates that many
ultramarathon runners attempt to achieve personal goals such as finishing or achieving a time,

20
instead of placement goals (Acevedo et al., 1992; Simpson et al., 2014). In addition to
striving for mastery goals, participants in the present study reported high self-efficacy and
perceived control, independent of whether they were in the experimental or control group.
These findings suggest that participants in this study typically experienced a challenge
response, rather than a threat response, to the upcoming ultramarathon. This challenge
response was accompanied by positively-toned emotions (cf. Jones et al., 2009); emotional
states before the start were characterised by excitement, happiness, and some anxiety. Self-
efficacy and perceived control values indicated that participants believed that they could cope
with event demands and finish the event (i.e., high self-efficacy) and that they perceived
themselves as having sufficient control over their responses to the event demands (i.e., high
perceived control). The self-talk intervention did not enhance these perceptions.
The presented data highlight the demands facing ultramarathon runners. Specifically,
the ultramarathon runners experienced increasing intensities of perceived effort, exercise-
induced pain (aching and burning muscles), and injury-related pain during the ultramarathon.
During actual endurance events such as the ultramarathon, endurance athletes also need to
cope with a plethora of additional stressors. These stressors include gastrointestinal problems,
thoughts about quitting, and adverse weather and conditions (Antonini Philippe et al., 2016;
Holt et al., 2014). In the present study, navigating in the dark was a key stressor experienced
by runners. These additional stressors are rarely experienced by participants performing in
sport science laboratories and non-competitive field settings, which is where the majority of
research on psychological interventions for endurance performance has been conducted
(McCormick et al., 2015). It is important that endurance athletes can cope with these
stressors, as they could elicit a potentially debilitative emotional response (e.g., Martinent &
Ferrand, 2009) and increase the demands of the situation compared to non-competitive
environments.

21
Future Research Suggestions
Although research examining the effects of psychological interventions at actual
endurance events is encouraged, this study highlights methodological challenges associated
with collecting data at an endurance event. Performance times in events such as the
ultramarathon may not compare to  performance times in other events, because
of differences in distance and race profile, and unique features of events. Randomised,
controlled, posttest-only designs may therefore be an appropriate choice of design, but these
designs require high participant numbers for statistical power, which may not be achievable.
Future experimental research at actual endurance events could use posttest-only designs to
examine performance in mass-participation events that involve thousands of entrants, or use
randomised, controlled, pretest-posttest designswhich have higher statistical powerto
examine performance in events that have less within-subject variability in performance times
between events. Such research is encouraged because of the plethora of additional stressors,
and associated emotions, experienced in actual events.
Participants assigned to the control group were given an alternative control treatment.
Providing control participants with an alternative treatment was intended to discourage them
from seeking the self-talk workbook (i.e., contamination), which could have reduced the
observed effect of the self-talk intervention. Further, similar intervention demands were
placed on control participants as the experimental group (e.g., time demands, completion of
logs as evidence of compliance), which can reduce bias associated with study dropouts (W.
Borg, 1984). The alternative control treatment involved practising concentration grids to
develop concentration. It is questionable, however, whether this control intervention had
similar perceived value to the self-talk intervention. Although participants in the control
group reported that improvements in concentration were important to them, only three of 14
participants in the control group were still using the concentration grid at the six-month

22
follow-up. In contrast, 12 of 14 participants in the self-talk group reported still using self-talk.
The self-talk group might therefore have valued the content of their workbook more. Further,
experimental research suggests that the concentration grid might not be efficacious at
improving concentration when it is used in isolation, rather than as part of a PST package,
without extensive practice (Greenlees, Thelwell, & Holder, 2006). Finally, although the
alternative control treatment targeted concentration rather than performance, an improvement
in concentration could potentially improve ultramarathon performance; that is, the targeted
outcomes are somewhat related (cf. W. Borg, 1984). When designing the concentration
workbook, care was therefore taken not to suggest performance-relevant cues that
participants could apply the concentration grid to during the ultramarathon; omitting this
information might have detrimentally affected the perceived value of the alternative control
treatment. Although using the concentration workbook as an alternative control treatment has
limitations, no participants in the control group reported being aware of the content of the
self-talk intervention. This suggests that the alternative control treatment prevented
contamination and therefore served a valuable purpose. Researchers conducting
psychological intervention research are encouraged to consider different ways of including an
alternative control treatment, to reduce sources of bias that could influence conclusions.
Limitations
Intake interviews were conducted before participants were allocated to experimental
and control groups. Once participants had received their workbook, the researcher did not
speak with participants about their particular workbook by video call or telephone. Instead,
participants were able to ask questions using a survey, and these questions were answered by
email. These decisions were made to prevent biasing the results by systematically introducing
expectation effects. This intervention format, however, is unrepresentative of how self-talk
interventions are delivered in applied settings (e.g., Latinjak, Font-Lladó, Zourbanos, &

23
Hatzigeorgiadis, 2016). Additional personalised support from a sport psychologist could have
encouraged even greater compliance, refined participants use of self-talk, and ultimately
increased the size of the intervention effect.
With consideration to external validity, participants in the study were those
ultramarathon entrants who volunteered for the research (22% of all entrants), rather than a
random selection of all entrants (cf. Bracht & Glass, 1968). The participants included in the
study may not therefore be representative of ultramarathon runners as a population. For
example, they may have had less competing time commitments, allowing them to meet the
time demands of the research. They may also have different attitudes towards sport
psychology. For example, they may have been less influenced by stigmas associated with
working with a sport psychologist, they may have had greater confidence in sport psychology
consultants, or they have had been more open to working with a sport psychologist (S. B.
Martin, Bochum, Lavallee, & Page, 2002). As a consequence, participants may have
perceived the intervention differently and been more likely to comply with the intervention.
Nevertheless, people who are willing and able to participate in studies could be more
representative of sport psychology consumers if common characteristics, such as time
availability and attitudes towards sport psychology, determine whether endurance athletes
volunteer for studies and pursue psychological assistance.
Conclusion
This study provides unique data on psychological intervention effects at an actual
endurance event, using a strong experimental design. The results showed that learning to use
strategic, motivational self-talk did not affect pre-event self-efficacy or perceived control. It
also did not affect performance in the ultramarathon, although this latter finding could be
explained by the sample size and variability in performance times. Nevertheless, the data
suggested that most participants found the intervention helpful and continued to use it six

24
months after their commitment to the research, particularly in endurance events and to a
lesser extent in training. Participants continued to use self-talk to cope with exertion, as well
as other stressors. Additional research examining the effects of psychological interventions
on performance in actual endurance events, where a plethora of stressors and associated
emotions are experienced, is encouraged.
References
Acevedo, E. O., Dzewaltowski, D. A., Gill, D. L., & Noble, J. M. (1992). Cognitive
orientations of ultramarathoners. The Sport Psychologist, 6, 242252.
http://doi.org/10.1123/tsp.6.3.242
Antonini Philippe, R., Rochat, N., Vauthier, M., & Hauw, D. (2016). The story of
withdrawals during an ultra-
courses of experience. The Sport Psychologist, 30, 361375.
http://doi.org/10.1123/tsp.2016-0039
Bandura, A. (2006). Guide for constructing self-efficacy scales. In F. Pajares & T. Urdan
(Eds.), Self-efficacy beliefs of adolescents (pp. 307337). Greenwich, CT: Information
Age.
Barwood, M. J., Corbett, J., Wagstaff, C. R. D., McVeigh, D., & Thelwell, R. C. (2015).
Improvement of 10-km time-trial cycling with motivational self-talk compared with
neutral self-talk. International Journal of Sports Physiology and Performance, 10, 166
171. http://doi.org/10.1123/ijspp.2014-0059
Batterham, A. M., & Hopkins, W. G. (2006). Making meaningful inferences about
magnitudes. International Journal of Sports Physiology and Performance, 1, 5057.
http://doi.org/10.1123/ijspp.1.1.50
Blanchfield, A. W., Hardy, J., de Morree, H. M., Staiano, W., & Marcora, S. M. (2014).
Talking yourself out of exhaustion: The effects of self-talk on endurance performance.

25
Medicine & Science in Sports & Exercise, 46, 9981007.
http://doi.org/10.1249/MSS.0000000000000184
Borg, G. A. (1998). Borg’s perceived exertion and pain scales. Champaign, IL: Human
Kinetics.
Borg, W. (1984). Dealing with threats to internal validity that randomization does not rule
out. Educational Researcher, 13(10), 1114.
http://doi.org/10.3102/0013189X013010011
Bracht, G. H., & Glass, G. V. (1968). The external validity of experiments. American
Educational Research Journal, 5, 437474. http://doi.org/10.3102/00028312005004437
Conroy, D. E., Elliot, A. J., & Hofer, S. M. (2003). A 2 X 2 Achievement Goals
Questionnaire for sport: Evidence for factorial invariance, temporal stability, and
external validity. Journal of Sport & Exercise Psychology, 25, 456476.
http://doi.org/10.1123/jsep.25.4.456

occurring muscle pain during exercise: assessment and experimental evidence. Medicine
& Science in Sports & Exercise, 29, 9991012. http://doi.org/10.1097/00005768-
199708000-00004
Greenlees, I., Thelwell, R., & Holder, T. (2006). Examining the efficacy of the concentration
grid exercise as a concentration enhancement exercise. Psychology of Sport and
Exercise, 7, 2939. http://doi.org/10.1016/j.psychsport.2005.02.001
Hatzigeorgiadis, A., & Biddle, S. J. H. (2008). Negative self-talk during sport performance:
Relationships with pre-competition anxiety and goal-performance discrepancies. Journal
of Sport Behavior, 31, 237253.
Hatzigeorgiadis, A., Galanis, E., Zourbanos, N., & Theodorakis, Y. (2014). Self-talk and
competitive sport performance. Journal of Applied Sport Psychology, 26, 8295.

26
http://doi.org/10.1080/10413200.2013.790095
Hatzigeorgiadis, A., Zourbanos, N., Latinjak, A., & Theodorakis, Y. (2014). Self-talk. In A.
Papaioannou & D. Hackfort (Eds.), Routledge companion to sport and exercise
psychology: Global perspectives and fundamental concepts (pp. 370383). London,
England: Taylor & Francis.
Holt, N. L., Lee, H., Kim, Y., & Klein, K. (2014). Exploring experiences of running an
ultramarathon. The Sport Psychologist, 28, 2235. http://doi.org/10.1123/tsp.2013-0008
Jones, M., Meijen, C., McCarthy, P. J., & Sheffield, D. (2009). A Theory of Challenge and
Threat States in Athletes. International Review of Sport and Exercise Psychology, 2,
161180. http://doi.org/10.1080/17509840902829331
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 Psychology,
27, 407431. http://doi.org/10.1123/jsep.27.4.407
Latinjak, A. T., Font-Lladó, R., Zourbanos, N., & Hatzigeorgiadis, A. (2016). Goal-directed
self-talk interventions: A single-case study with an elite athlete. The Sport Psychologist,
30, 189194. http://doi.org/10.1123/tsp.2015-0120
Malek, M. H., Housh, T. J., Berger, D. E., Coburn, J. W., & Beck, T. W. (2004). A new
nonexercise-based VO2max equation for aerobically trained females. Medicine &
Science in Sports & Exercise, 36, 18041810.
http://doi.org/10.1249/01.MSS.0000142299.42797.83
Malek, M. H., Housh, T. J., Berger, D. E., Coburn, J. W., & Beck, T. W. (2005). A new non-
exercise-based VO2max prediction equation for aerobically trained men. Journal of
Strength and Conditioning Research, 19, 559565. Retrieved from
http://journals.lww.com/nsca-jscr
Martin, G. L., Vause, T., & Schwartzman, L. (2005). Experimental studies of psychological

27
interventions with athletes in competitions: Why so few? Behavior Modification, 29,
616641. http://doi.org/10.1177/0145445503259394
Martin, S. B., Bochum, R., Lavallee, D., & Page, S. J. (2002). Development and
psychometric evaluation of the Sport Psychology Attitudes - Revised Form: A multiple
group investigation. The Sport Psychologist, 16, 272290.
http://doi.org/10.1123/tsp.16.3.272
Martinent, G., & Ferrand, C. (2009). A naturalistic study of the directional interpretation
process of discrete emotions during high-stakes table tennis matches. Journal of Sport &
Exercise Psychology, 31, 318336. http://doi.org/10.1123/jsep.31.3.318
McCormick, A. (2016). Psychologically-informed methods of enhancing endurance
performance (Doctoral thesis, University of Kent, Kent, England). Retrieved from
https://kar.kent.ac.uk/57284/
McCormick, A., Meijen, C., & Marcora, S. (2015). Psychological determinants of whole-
body endurance performance. Sports Medicine, 45, 9971015.
http://doi.org/10.1007/s40279-015-0319-6
McCormick, A., Meijen, C., & Marcora, S. (2016). Psychological demands experienced by
recreational endurance athletes. International Journal of Sport and Exercise Psychology.
http://doi.org/10.1080/1612197X.2016.1256341
Miller, A., & Donohue, B. (2003). The development and controlled evaluation of athletic
mental preparation strategies in high school distance runners. Journal of Applied Sport
Psychology, 15, 321334. http://doi.org/10.1080/714044200
Page, J., & Thelwell, R. (2013). The value of social validation in single-case methods in sport
and exercise psychology. Journal of Applied Sport Psychology, 25, 6171.
http://doi.org/10.1080/10413200.2012.663859
Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and quasi-

28
experimental designs for generalized causal inference. Boston, MA: Houghton Mifflin.
Simons, J. (2012). Endurance psychology. In I. Mujika (Ed.), Endurance training – Science
and practice (pp. 201210). Vitoria-Gasteiz, Basque Country: Iñigo Mujika S.L.U.

The Sport Psychologist, 28, 176185.
http://doi.org/10.1123/tsp.2013-0064
Terry, P. C., Lane, A. M., Lane, H. J., & Keohane, L. (1999). Development and validation of
a mood measure for adolescents. Journal of Sports Sciences, 17, 861872.
http://doi.org/10.1080/026404199365425
Van Raalte, J. L., Morrey, R. B., Cornelius, A. E., & Brewer, B. W. (2015). Self-talk of
marathon runners. The Sport Psychologist, 29, 258260.
http://doi.org/10.1123/tsp.2014-0159
Wallace, P. J., McKinlay, B. J., Coletta, N. A., Vlaar, J. I., Taber, M. J., Wilson, P. M., &
Cheung, S. S. (2017). Effects of motivational self-talk on endurance and cognitive
performance in the heat. Medicine & Science in Sports & Exercise, 49, 191199.
http://doi.org/10.1249/MSS.0000000000001087
Notes
1 Additional detail on the methods is available (McCormick, 2016).
2 All event competitors carried GPS tags that gave real-time information on their progress.
           were only
available for the second year. There was insufficient GPS and self-report data to compare
ratings of perceived exertion and pain, relative to pace, between conditions.

29
Figures
Figure 1. Overview of the study timeline.
010 20 30 40 50 60
6
8
10
12
14
16
18
20 A
Distance (miles)
RPE
010 20 30 40 50 60
0
2
4
6
8
10
Exercise-induced pain
Injury-related pain
B
Distance (miles)
Pain
Figure 2. Mean and standard deviation ratings of perceived exertion (RPE) (panel A) and
ratings of pain (panel B) during the ultramarathon.

30
Tables
Table 1
Sport Emotions and Fatigue Before the Ultramarathon and After Completing
the Ultramarathon
Pre-event
Post-event
Anxiety
M = 1.50, SD = 0.92
M = 0.067, SD = 0.13
Dejection
M = 0.13, SD = 0.37
M = 0.29, SD = 0.53
Excitement
M = 2.52, SD = 0.82
M = 2.25, SD = 1.01
Anger
M = 0.056, SD = 0.20
M = 0.17, SD = 0.27
Happiness
M = 2.39, SD = 0.98
M = 2.96, SD = 0.93
Fatigue
M = 0.66, SD = 0.59
M = 2.25, SD = 0.94
Note. Anxiety, dejection, excitement, anger, and happiness were measured
using the Sport Emotion Questionnaire (Jones, Lane, Bray, Uphill, & Catlin,
2005). Fatigue was measured using the Brunel Mood Scale (Terry, Lane,
Lane, & Keohane, 1999). Post-event values were reported retrospectively, a
mean of 4 days (SD = 6) after completing the event. Scores can range from 0
to 4.