Success in an athletic event involves ongoing conscious and unconscious assessment of the immediate and anticipated internal and external environments. Athletes are constantly comparing how fatigued they feel at any given point in an event to how fatigued they expected to feel at that point; this perception is compared with a preset “template” that they have subconsciously created beforehand (16). They are then forced to make a decision as to whether they should increase or decrease their pace, based on how much physiological disturbance they are willing to accept. Ulmer (27) first coined the term “teleoanticipation” based on work that he performed on athletes who were asked to exercise and maintain certain ratings of perceived exertion (RPE) levels (6), allowing them to adjust their pacing accordingly. Ulmer concluded that there is likely a “regulation center” in the brain that performs precise calculations about metabolic reserves, as well as time and effort needed to finish exercise bouts at a given intensity. Ulmer's study on athletes corroborated results from a previous experiment by Vidacek and Wishner (29) on the effect of task length on muscle activity. The authors showed that even during a simple task such as holding a weight off the floor, energy sources are conserved when the task duration is unknown. In runners, conservation of energy is crucial to reaching the endpoint of a training run or a race. If the endpoint is not known by an individual, it is likely that energy output and RPE would be low compared with a condition in which the endpoint is in fact known. This is done in an attempt to ensure that the bout of exercise is finished in full, holding resources in reserve in case the duration is beyond a faster pace's capacity.
The majority of studies using a laboratory treadmill have implemented fixed pacing, usually to a speed that is relative to the subject. Self-pacing studies have an arguably higher level of ecological validity because running events are typically overground, and athletes are able to adjust pace as needed. When the distance of an event is known and subjects are able to self-pace, they choose an optimal speed to complete the event at the highest level of efficiency (9). World records in running events from 1500 m up to the marathon have had a variation in running speed of 1–5%, showing that elite runners constantly adjust their pace during a race to optimize performance (4). In 2006, Billat et al. (5) demonstrated that a “freely” or self-paced run elicited significantly higher
values, heart rate (HR), and blood lactate concentration compared with a fixed pace run over 10,000 m. Garcin et al. (13) showed that subjects were able to maintain a significantly faster velocity in a self-paced compared with a fixed pace run, whereas there were no differences in RPE or oxygen uptake. A novel self-paced, fixed time cycling
protocol was introduced by Mauger and Sculthorpe (20). When compared with a standard incremental
protocol where subjects did not know the endpoint and were asked to run to volitional exhaustion, the 10-minute self-paced protocol resulted in both higher peak
values and power outputs.
Tasks with unknown endpoints have been investigated to some degree in the sports science literature and can provide researchers with useful insight into the sources and management of fatigue. Baden et al. (2) showed that when subjects were asked to run at a fixed pace with an unknown endpoint, they consumed significantly less oxygen than when running with a known endpoint, showing that they were more economical in the former condition compared with the latter. Their RPE values were similar between conditions. Other studies have shown that knowledge of endpoint can affect perception of fatigue. Coquart and Garcin (7) asked subjects to run at a fixed pace with an unknown endpoint until exhaustion. They were asked to run at that same pace 2 more times, the same distance that was reached and for the same amount of time. For the last 2 trials, the subjects were aware of the endpoint, which was a set distance in trial 2 and a set time in trial 3. Subjects' RPE was higher and their HR was slightly lower when the endpoint was known even though they were running at the same speed, showing that perceived exertion is affected by knowledge of an endpoint. In addition, recent studies have shown scalar linear properties of RPE during competition and simulated competition (12,16). Because level of enjoyment or affect has been shown to influence RPE during exercise (22), Baden et al. (2) also assessed affect by the Feeling Scale (FS) throughout each of their experimental conditions, including a third condition in which deception was used; subjects were asked to run an additional 10 minutes at the end of the planned 10 minutes. They found that at the 11-minute point in the condition where the subjects were asked to continue running, the reported affect was significantly less positive than the other 2 conditions.
Perceived exertion has also been shown to be influenced by focus of attention, that is, cognitive strategies of association or dissociation. Associative thoughts are those in which the focus is placed on cues received from the body, whereas dissociative thoughts are those in which a person does not tune into bodily cues and engages in thoughts such as daydreaming. There is good evidence that at greater levels of intensity, physiological cues dominate attentional focus (25). Increased levels of competition and motivation, which generally arise from more organized training and improved cardiovascular fitness, have been shown to be associated with increased use of associative cognitive strategies (19,21,23,24). The expected task duration affects pacing strategy, which likely affects perceived exertion and the direction of attentional focus.
The purpose of this study was to examine the effects of self-paced running with an unknown endpoint vs. a known endpoint on HR, RPE, attentional focus, and affect. It was hypothesized that subjects would run significantly faster in the known endpoint condition, leading to differences between the conditions in HR, RPE, and affect. Likewise, it was also hypothesized that associative cognitive strategies would be used more in the known endpoint compared with the unknown endpoint condition.
Experimental Approach to the Problem
The purpose of this study was to examine the effect of endpoint knowledge on psychophysiological variables during self-paced running. Experienced recreational male and female runners were used as subjects for this study. Each subject completed 3 separate visits to the laboratory with a minimum of 48 hours between each test. The first visit consisted of an incremental treadmill test to exhaustion to determine their
. The second visit was a run with an endpoint that was not disclosed to the subject until the end of the session. The third and final visit was a run to the same distance that the subject completed in the second visit. Detailed information on the 3 visits is presented below.
Twenty-two recreational runners (11 men and 11 women) volunteered to participate in this study. All subjects provided written informed consent before testing. The study was approved by the Institutional Review Board of the university's Office of Responsible Research Practices. Descriptive characteristics of the subjects are presented in Table 1. All subjects were considered “low risk” according to the American College of Sports Medicine risk stratification guidelines (26). They were also required to have run a minimum of 10 miles per week for the previous 6 months and be free of any musculoskeletal injuries.
Rating of Perceived Exertion
The Borg 15-point scale was used to assess perceived exertion and ranged from 6 to 20, with verbal anchors such as “very light,” “somewhat hard,” and “very hard” (6). The scale was developed out of a need to quantify the perception of effort and has been widely useful and applicable across sex, age, and nationality.
Rating of perceived exertion represents a host of sensations related to the physical strain of exercise, but it does not address the affective responses to exercise intensity and situation. For example, a person may report an RPE value of 17 (corresponding to “very hard”) but feel “very good” or “very bad” at that level of exertion. Therefore, in light of this reason and the findings that psychological variables play a very important role in exercise and sport, an additional scale to measure affect was used along with RPE (14,28). The FS was developed to assess fluctuations in affect across time during exercise to determine when and if exercise is pleasurable or unpleasant to individuals (14). The FS is a single-item measure with response choices on an 11-point scale ranging from +5 (very good) to −5 (very bad) with zero as a neutral midpoint.
The level of attentional focus, as measured by the reported percentage of dissociative thoughts, was assessed. The method used was similar to that used by Baden et al. (3), through use of a bipolar line with “associative” at one end and “dissociative” at the other end, where subjects are asked to mark an “X” on the line corresponding with their level of association (thoughts directed at bodily symptoms such as breathing rate, HR, sweating, or pain) or dissociation (external thoughts that are distracting from exercise, such as daydreaming or thinking about activities they may be participating in at a later time) (3,17). However, rather than asking subjects to mark an “X” while they are running, they were asked to verbally state their current percentage of dissociative thoughts.
Laboratory Visit 1
The consent form was reviewed and any questions were addressed before securing the subjects' written consent. The subjects then reported to the exercise testing laboratory for a
test to assess their cardiorespiratory fitness level. The 15-point RPE scale was described in detail before testing and the subjects were asked to report RPE during the last 30 seconds of each 2-minute stage (6). A modified Astrand-Saltin
protocol using incremental increases in grade was used (15). Subjects were fitted with a Polar HR chest strap and monitor (Polar Electro Oy, Kempele, Finland) and asked to warm up at a light-to-moderate intensity on the laboratory treadmill for 5 minutes before beginning the test. The subjects were then fitted with the appropriate headgear and mouthpiece and began the
test by running for 2 minutes at a pace slightly faster than his or her normal training pace and 0% grade. Every 2 minutes thereafter, the speed was kept constant and the grade increased by 2% until they reached volitional exhaustion. The 2 highest consecutive relative
values, in milliliter per kilogram per minute, were averaged to obtain
. Sampling frequency of the metabolic cart was 0.067 Hz.
Tests were considered “true” maximal tests if 2 of the following 3 criteria were obtained: (a) RPE: equal to or greater than 18, (b) HR: within 10 b·min−1 of the subject's age predicted maximal HR (calculated with the following formula: 220 − age = HRmax), (c)
plateau: the difference between the peak relative
value and the value in the immediately preceding 15 s of 2.0 mL·kg−1·min−1 or less.
Laboratory Visit 2—Unknown Endpoint
Subjects were fitted with a Polar HR monitor (chest strap) and were asked to warm up by running on the laboratory treadmill at a constant self-selected fixed pace for 5 minutes. After this initial warm-up, they were stopped briefly so that the treadmill distance could be zeroed. The subjects were then asked to run for an unspecified distance, which was calculated after the first laboratory visit, and to put forth a “good effort.” They were told that the distance was based on their running history and average weekly or daily mileage and was not longer than their longest run in the previous 6 months. For example, if a subject reported running 5 miles per day for 4 days a week for the last 6 months with a long run of 12 miles, the distance for this condition would be 75% of their normal daily run or 5 × 0.75 = 3.75 miles. If the subject reported running a range, such as 5–7 miles per day for 3–5 days per week, the average of the range was taken; this subject would run (5 + 7)/2 × 0.75 = 4.5 miles. The RPE, FS, and HR were all assessed at varying time points during the testing ranging from 1 to 4 minutes apart. Attentional focus was assessed approximately every 5–6 minutes. Subjects were told that they would be asked to report their level of exertion, feelings, and percent dissociation at random time points to make them aware that they would not be able to count the number of assessments and estimate the time they had been running. At no point during the testing were subjects told either the cumulative distance or time. When they reached the predetermined but undisclosed endpoint, they were told that they had completed the trial.
Laboratory Visit 3—Known Endpoint
Randomization to treatment order was not possible because it was necessary to have the subjects be unaware of the distance that they would complete in the first experimental trial (visit 2—unknown endpoint). Therefore, the third laboratory visit was identical to the second visit, with the exception that subjects were aware of the distance that they were asked to run. They were still, however, blind to the elapsed time, speed, and distance during the course of the run. They were asked to “complete the same distance as in visit 2” and were instructed to warm up for 5 minutes at the same self-selected speed chosen in visit 2. Laboratory visits 2 and 3 were scheduled for approximately 1.5–2.0 hours.
Descriptive statistics were calculated and presented for all subjects; results from the
testing were calculated separately for men and women. The significance level for all analyses was set a priori at p ≤ 0.05 and effect size using partial eta-square was reported for repeated-measures analyses of variance (ANOVAs). Cohen's d was provided as an effect size for paired samples t-tests, and R2 was given as an effect size for simple regression analyses. The effect sizes are given in addition to p values to provide information about the differences between groups and any practical significance.
The subjects reported running 29.02 miles per week on average (SD = 16.87). The mean required distance was 4.15 miles (SD = 1.17, range 3.0–7.5). The mean time to complete the required distance in the unknown endpoint condition was 32:49.0 (SD = 8:01.3) minutes and in the known endpoint condition was 31:20.1 (SD = 8:02.5) minutes. This difference was statistically significant, t(20) = 3.915, p = 0.001, d = 0.865.
Ratings of Perceived Exertion
The final reported RPE for each condition was averaged and compared. The difference in final RPE between the unknown (mean = 14.9, SD = 1.8) and the known (mean = 15.5, SD = 2.0) endpoint conditions was not significant, t(21) = −1.914, p = 0.069. The mean RPE for each subject was also calculated and a paired samples t-test showed that the average RPE for the unknown endpoint condition was 13.5 (SD = 1.4) and for the known endpoint condition was 13.8 (SD = 1.8); this difference was also not significant, t(21) = −1.130, p = 0.271.
Because there was a high level of variability in the amount of time the subjects ran and RPE was assessed at absolute time points, there was also variation in the number of data points for each subject (Figure 1). Therefore, the ratings of perceived exertion were normalized to the percentage of time complete, 10% increments and a range from 10 to 100% of the time complete, and analyzed using a 2 (endpoint condition) × 10 (percent time increments) repeated-measures ANOVA (Figure 2). There was not a significant main effect of endpoint condition (F(1,20) = 0.069, p = 0.796,
= 0.003) or an interaction between time and condition, F(7,14) = 1.000, p = 0.471,
= 0.333. There was an effect of time, which was significant at both the linear (F(1,20) = 104.655, p < 0.001,
= 0.840) and the quadratic levels (F(1,20) = 33.426, p < 0.001,
= 0.626). The assumption of sphericity was not met, so the Greenhouse–Geisser adjustment was made for the main effect of time, F(2.640,12) = 52.110, p ≤ 0.001,
= 0.723. All other assumptions of repeated-measures ANOVA were met.
The mean HR over the entire run with an unknown endpoint was 166.8 (SD = 14.7) b·min−1. The mean HR during the known endpoint run was 169.7 (SD = 14.7) b·min−1. These were not significantly different, t(21) = −1.451, p = 0.161. When the final HR measurement for each condition was averaged, there was also not a difference between the unknown endpoint (mean = 173.5, SD = 13.9) and the known endpoint conditions (mean = 176.0, SD = 14.7) using a paired samples t-test, t(21) = −1.552, p = 0.136.
The average FS values over the course of each run were 2.164 (SD = 1.602) for the unknown and 2.124 (SD = 1.595) for the known endpoint condition. This difference was not significant, t(21) = 0.143, p = 0.887. The reported feelings declined over the course of each run, regardless of endpoint condition. The average last reported FS scores were 0.64 (SD = 2.38) for the unknown endpoint condition and 0.73 (SD = 2.43) for the known endpoint condition. This difference was not significant, t(21) = −0.245, p = 0.809.
The final assessment of attentional focus (percent dissociative thoughts) for each subject was averaged for each endpoint condition. The mean final value for the unknown endpoint was 56.0% (SD = 18.1) and was 45.2% (SD = 21.0) for the known endpoint condition. A paired samples t-test showed that the difference between these 2 means was significant, t(20) = 2.275, p = 0.034, d = 0.499.
A simple linear regression was used with the final attentional focus assessment as the dependent variable and
as a continuous independent variable. Results showed that
was not able to significantly predict percent dissociative thoughts in the unknown endpoint condition (t = −0.684, p = 0.502). When the average percentage over the entire run was used as the dependent variable,
was also not a significant predictor (t = −1.335, p = 0.198). However,
was a significant predictor of the final report of attentional focus in the known endpoint condition and explained 31.7% of the variance (t = −2.972, p = 0.008, R2 = 0.317).
was also a significant predictor of the average attentional focus over the entire known endpoint run (t = −2.439, p = 0.025, R2 = 0.239). Further examination of collinearity statistics revealed no problems with collinearity, and upon examination of the residuals, there were no violations of the assumptions of linear regression.
The last assessment of attentional focus (percent dissociative thoughts) was significantly negatively correlated to RPE levels at both 90% (r = −0.668, p = 0 0.005, ρ = 0.667) and 100% (r = −0.572, p = 0 0.021, ρ = 0.572) of the completed time in the known endpoint condition. The greater the proportion of dissociative thoughts, the lower the RPE near the end of the run. In the unknown endpoint condition, attentional focus was not correlated with RPE at either the 90% (r = −0.324, p = 0.221) or 100% (r = −0.286, p = 0.283) of the completed time.
The main objective of this study was to investigate how the knowledge of an endpoint affects different psychophysiological variables, when the subjects are able to use a self-selected pace throughout testing. To the author's knowledge, this is the first study to implement an unknown endpoint and the ability to self-pace. All other studies using self-pacing protocols have been those in which subjects cycle or run at a very high intensity, during a race (or simulated race), or “to exhaustion,” when they decide that they cannot go any farther. One important finding from this study was that overall and final RPE values were not significantly different between endpoint conditions despite a significantly faster completion time in the known endpoint condition. Even when RPE values were normalized to the percentage of time completed, there was not a significant difference between the endpoint conditions. This shows that when the endpoint was known, the subjects ran faster at the same level of perceived exertion.
It was hypothesized that the subjects would run significantly faster when they were aware of the endpoint than during the unknown endpoint condition. This did occur, even though the subjects received no feedback during either condition. When the subjects were not aware of the distance that they would be running, they ran more slowly than when they were aware of the distance. Based on the concept of teleoanticipation, the subjects conserved their metabolic resources when the endpoint was unknown because they did not want to fatigue before they reached the end of the trial.
Coquart and Garcin (7) asked their 14 subjects (all men) to run in 3 different conditions. The first consisted of a run with an unknown endpoint, in which the subjects were instructed to run to exhaustion. For the second run, the subjects were asked to run to the same distance that they covered as part of the first run to exhaustion. Finally, during the third run, the subjects were told to run to the same distance that was reached on the first run. The conditions, then, were “unknown endpoint,” “known duration,” and “known distance.” The researchers assessed RPE throughout the trials and discovered that RPE was significantly (p = 0.023) lower in the unknown endpoint condition compared with the other 2 conditions in which the distance or duration were known. For all conditions, the subjects were not able to self-pace; they were asked to run at a fixed pace coinciding with 90% of their maximal aerobic velocity. The authors found significant differences in RPE at the 40, 60, and 80% time points but not at the 20% time point. When using paired samples t-tests to compare the 20, 40, 60, and 80% relative time points between known and unknown endpoint conditions in our subjects, there were no differences between any of the time points (all p > 0.05). In the current study, there were also no significant differences in average or maximum RPE values between unknown and known endpoint conditions. However, our subjects were able to adjust their pace throughout the trials and ran faster in the known endpoint condition. The ability to self-pace may have enabled our subjects to manage the effort they expended consistently regardless of knowledge of remaining distance.
Faulkner et al. (11) used inaccurate, accurate, or no distance feedback during 4 self-paced 6-km treadmill runs. The “inaccurate” feedback was further split into either premature or delayed feedback, so each subject completed the 6-km run a total of 4 times. Their subjects consisted of 13 healthy, physically active young men. They asked their subjects to complete the trials as quickly as they could. They found that completion times were significantly (p < 0.001) slower when the subjects were given no feedback about the distance that they had covered. Performance was unaffected by inaccurate feedback, which was also shown by another study of 15 male cyclists (11). Faulkner et al. also found that RPE was similar between all conditions, further supporting the idea that an RPE template is created before the trial based on distance. Additionally, only in the 3 conditions in which feedback was given (either accurate or inaccurate) was the “end spurt” phenomenon clearly observed. Because a key concept of teleoanticipation is based on how much time or distance there is left to complete, feedback is an important determinant in how metabolic sources are used during exercise. In the present study, even in the “known endpoint” condition, there was no feedback given to the subjects. They were blind to all feedback and were not told how fast they were running, how much time had elapsed, or how far they had gone. They simply knew the distance in which they would be told they could stop running. Consequently, they had an RPE template, although less precise, for the known endpoint that was based on the quality of their recall of the previous exercise bout. This RPE template was also likely to be influenced by level of experience; the subjects with greater running and racing history may have been more mindful of their pace and effort level required to complete the distance most efficiently.
Despite the fact that the subjects ran at an overall faster pace when the endpoint was known, the difference between the 2 HR means was not significant (p = 0.161). During previous studies of self-paced running, there are often no differences in HR between conditions. In the study by Faulkner et al. (12), they did not find any differences in absolute average HR between 2 very different races (7-mile simulated race and a 13.1-mile half-marathon). They also did not find any differences between the final recorded HR during the 7-mile race (mean = 186.7, SD = 9.7) and the half-marathon (mean = 185.8, SD = 6.8). When the recorded HR data were normalized to percentage of the subjects' maximum HR, there were still no differences between conditions. This is despite the subjects running nearly twice as long in the half-marathon and running at a significantly faster pace in the first 7 miles of the 7-mile race compared with the first 7 miles of the half-marathon. In this study, when the HR was normalized to the percentage of each subject's maximum HR obtained from the
test, the mean of the unknown endpoint condition was 87.7% (SD = 5.2%) and for the known endpoint was 89.2% (SD = 5.7%). These were not significantly different (p = 0.159).
It was hypothesized that the subjects would have significantly more negative affect during the known endpoint condition, based on the assumption that they would be running at an overall faster pace that could be unpleasant and result in reports of less positive affect. It was discovered that there were no significant differences in average FS scores between the unknown (mean = 2.164, SD = 1.602) and known (mean = 2.124, SD = 1.595) endpoint conditions. After comparing the last reported FS score, there were still no differences between endpoint conditions. The mean final FS score in the unknown endpoint condition was 0.64 (SD = 2.38) and in the known endpoint condition was 0.73 (SD = 2.43). In both endpoint conditions, there were very large SDs in the final reported FS score. This shows that across endpoint conditions, the subjects showed a large amount of variation in their affect. Further analysis of these data showed that the range of final FS scores in the unknown endpoint condition was from −4 to +4, and in the known endpoint condition was −5 to +5. Additionally, the median final FS score in the unknown endpoint condition was 0, and was +1 in the known endpoint condition. These descriptive statistics help to illustrate the fact that although some subjects felt more negative toward the end of exercise, others felt surprisingly good. One explanation for the FS variability could be differences in personality; for example, extraverts have been found to have better mood scores than introverts at high intensities (10) and all 5 dimensions of personality are shown to be related to exercise behavior in some form (8). In addition to personality, variables such as age, sex, and body composition could all be moderating factors on affect during running under known and unknown endpoint conditions and warrants further investigation.
It was found that fitness was not a significant predictor of attentional focus in the unknown endpoint condition (t = −0.684, p = 0.502). However, in the known endpoint condition, greater fitness levels did correspond with higher levels of reported associative thoughts (t = −2.972, p = 0.008, R2 = 0.317). Masters and Lambert (18) found that marathoners reported that roughly 75% of their thoughts during a marathon race were associative. In addition, a study of runners competing in the Olympic Trials Marathon found that top finishers used a greater percentage of associative thoughts compared with slower finishers (23). Other studies have shown that as competition increases, there is also a corresponding move toward more associative thoughts (21,24), and it is believed that highly competitive and motivated athletes use associative thoughts to enhance performance (19). During the known endpoint run, the subjects in the current study may have been competing with their unknown endpoint run, especially those who had relatively higher fitness levels.
A limitation to the current study is that subjects may not have used the ability to change speed as much as they would have if increasing or decreasing simply meant speeding up or slowing down at a subconscious level. The actual pressing of a button to change speed may not have been performed by subjects because they wanted to stay at a certain pace for as long as they could. Even slight variations in pace occur when running outdoors at a target pace. To create a more realistic self-pacing condition, future testing should be performed in a situation where subjects are not required to press a button to increase or decrease speed, such as an indoor track or a self-pacing treadmill. This would increase variability in pace, which is regularly seen in outdoor running. Another limitation to this study is that randomization into conditions was not used, which could potentially lead to a learning effect. Finally, because of the low sample size, the analyses may have been underpowered.
The findings of this study could be useful for runners and coaches. We showed that there were no differences in perceived exertion, affect, or HR between a self-paced run to either a known or unknown endpoint. However, when the endpoint was not known, the subjects chose a significantly slower pace overall. Perceiving the same level of exertion while running slower implies that not knowing when an exercise bout will be over can contribute to effort sense and should be examined in novice runners to further observe psychological and situational contributions to pacing and perceived exertion. Although fairly uncommon, some coaches will not disclose the distance that their athletes will be covering on a given day. This study shows that method to likely be ineffective at decreasing perceived exertion, affect, or HR. The study also showed that there are marked differences in attentional focus between endpoint conditions and that cardiorespiratory fitness can help to predict the choice of cognitive strategy used during exercise. Along with cardiovascular and strength training, perhaps a form of “brain training” could be implemented where runners practice using associative cognitive strategies in an attempt to increase overall performance. Further research and practice is needed to determine whether cognitive strategies used by runners can be modified through coaching in an attempt to enhance performance or affect.
This research was not funded and the authors have no conflicts of interest to declare.
1. Albertus Y, Tucker R, St Clair Gibson A, Lambert EV, Hampson DB, Noakes TD. Effect of distance feedback on pacing strategy and perceived exertion during cycling. Med Sci Sports Exerc 37: 461–468, 2005.
2. Baden DA, McLean TL, Tucker R, Noakes TD, Gibson ASC. Effect of anticipation during unknown or unexpected exercise duration on rating of perceived exertion, affect, and physiological function. Br J Sports Med 39: 742–746, 2005.
3. Baden DA, Warwick-Evans L, Lakomy J. Am I nearly there? The effect of anticipated running distance on perceived exertion and attentional focus. J Sport Exerc Psychol 26: 215–231, 2004.
4. Billat VL, Slawinski J, Danel M, Koralsztein JP. Effect of free versus constant pace on performance and oxygen kinetics in running. Med Sci Sports Exerc 33: 2082–2088, 2001.
5. Billat VL, Wesfreid E, Kapfer C, Koralsztein JP, Meyer Y. Nonlinear dynamics of heart rate and oxygen uptake in exhaustive 10,000 m runs: Influence of constant vs. freely paced. J Physiol Sci 56: 103–111, 2006.
6. Borg GA. Psychophysical bases of perceived exertion. Med Sci Sports Exerc 14: 377–381, 1982.
7. Coquart JB, Garcin M. Knowledge of the endpoint: Effect on perceptual values. Int J Sports Med 29: 976–979, 2008.
8. Courneya KS, Hellsten LAM. Personality correlates of exercise behavior, motives, barriers and preferences: An application of the five-factor model. Pers Indiv Differ 24: 625–633, 1998.
9. de Paula Viveiros J, Amorim FT, Alves MN, Passos RL, Meyer F. Run performance of middle-aged and young adult runners in the heat. Int J Sports Med, 33: 211–217, 2011.
10. Ekkekakis P, Hall EE, Petruzzello SJ. Some like it vigorous: Measuring individual differences in the preference for and tolerance of exercise intensity. J Sport Exerc Psychol 27: 350, 2005.
11. Faulkner J, Arnold T, Eston R. Effect of accurate and inaccurate distance feedback on performance markers and pacing strategies during running. Scand J Med Sci Sports 21: e176–e183, 2011.
12. Faulkner J, Parfitt G, Eston R. The rating of perceived exertion during competitive running scales with time. Psychophysiology 45: 977–985, 2008.
13. Garcin M, Danel M, Billat V. Perceptual responses in free vs. constant pace exercise. Int J Sports Med 29: 453, 2008.
14. Hardy CJ, Rejeski WJ. Not what, but how one feels: The measurement of affect during exercise. J Sport Exerc Psychol 11: 304–317, 1989.
15. Hawkins MN, Raven PB, Snell PG, Stray-Gundersen J, Levine BD. Maximal oxygen uptake as a parametric measure of cardiorespiratory capacity. Med Sci Sports Exerc 39: 103–107, 2007.
16. Joseph T, Johnson B, Battista RA, Wright G, Dodge C, Porcari JP, de Koning JJ, Foster C. Perception of fatigue during simulated competition. Med Sci Sports Exerc 40: 381–386, 2008.
17. Lima-Silva A, Silva-Cavalcante M, Pires F, Bertuzzi R, Oliveira R, Bishop D. Listening to music in the first, but not the last 1.5 km of a 5-km running trial alters pacing strategy and improves performance. Int J Sports Med, 33: 813–818, 2012.
18. Masters KS, Lambert MJ. The relations between cognitive coping strategies, reasons for running, injury, and performance of marathon runners. J Sport Exerc Psychol 11: 161–170, 1989.
19. Masters KS, Ogles BM. The relations of cognitive strategies with injury, motivation, and performance among marathon runners: Results from two studies. J Appl Sport Psychol 10: 281–296, 1998.
20. Mauger AR, Sculthorpe N. A new VO2
max protocol allowing self-pacing in maximal incremental exercise. Br J Sports Med 46: 59–63, 2012.
21. Okwumabua T. Psychological and physical contributions to marathon performance: An exploratory investigation. J Sport Behav 8: 163–171, 1985.
22. Rejeski WJ, Ribisl PM. Expected task duration and perceived effort: An attributional analysis. J Sport Psychol 2: 227–236, 1980.
23. Silva J, Appelbaum M. Association-dissociation patterns of United States Olympic marathon trial contestants. Cognit Ther Res 13: 185–192, 1989.
24. Summers JJ, Sargent GI, Levey AJ, Murray KD. Middle-aged, non-elite marathon runners: A profile. Percept Mot Skills 54: 963–969, 1982.
25. Tenenbaum G. A Social-Cognitive Perspective of Perceived Exertion and Exertion Tolerance. New York, NY: Wiley, 2001.
26. Thompson WR, Gordon NF, Pescatello LS. ACSM's Guidelines for Exercise Testing and Prescription. Philadelphia, PA: Lippincott Williams & Wilkins, 2009.
27. Ulmer HV. Concept of an extracellular regulation of muscular metabolic rate during heavy exercise in humans by psychophysiological feedback. Experientia 52: 416–420, 1996.
28. Vallerand R. Emotion in sport: Definitional, historical and social psychological perspectives. Cognitive Sport Psychology. W. Straub and J. Williams, Editors. 1984, Sport Sciences Associates: Lansing, NY. p. 65–78.
29. Vidacek S, Wishner J. Influence of expectation of task duration on efficiency of muscular activity. J Appl Psychol 55: 564–569, 1971.