Predicted and Actual Exercise Discomfort in Middle School Children


Medicine & Science in Sports & Exercise:
doi: 10.1249/MSS.0b013e3181c3aa62
Applied Sciences

Purpose: The purpose of this study was to use a match-mismatch paradigm to examine children's exercise discomfort during an aerobic shuttle run.

Methods: Thirty-four middle school females (n = 18) and males (n = 16) aged 11-14 yr participated. An Exercise Discomfort Index (EDI) was calculated as a rating of perceived exertion for the overall body (Children's OMNI Scale) × a rating of perceived muscle hurt (Children's OMNI Muscle Hurt Scale). Measurements were obtained immediately before (i.e., predicted) and after (i.e., actual) performance of the nationally standardized Progressive Aerobic Cardiovascular Endurance Run (PACER) shuttle test of aerobic fitness. Self-report physical activity and sport participation history were obtained before PACER performance.

Results: Two-way ANOVA (gender × assessment time point) showed a significant main effect for assessment time point: predicted EDI (means ± SD = 25.9 ± 20.1) was greater than actual EDI (means ± SD = 19.4 ± 17.8) for the total group (P = 0.021). However, neither the main effect of gender nor the gender × assessment time point interaction was significant. Idiographic analysis showed that overpredictors of discomfort reported less time (5.25 median h·wk−1) and engaged in less recreational activity than underpredictors (11.14 median h·wk−1). However, no significant relation (P = 0.508) was observed between PACER laps completed and exercise discomfort.

Conclusions: The sample of middle school children in this study predicted greater exercise discomfort than actually experienced when performing a PACER test. It is possible that a discomfort construct plays an important role in the initiation and maintenance of children's aerobic exercise, providing a basis for physical activity interventions.

Author Information

1Department of Nursing, University of Pittsburgh, Pittsburgh, PA; 2Department of Health & Physical Activity, University of Pittsburgh, Pittsburgh, PA; 3Falk School, University of Pittsburgh, Pittsburgh, PA; 4Department of Pathology and Psychiatry, University of Pittsburgh, Pittsburgh, PA; and 5Office of Measurement & Evaluation of Teaching, University of Pittsburgh, Pittsburgh, PA

Submitted for publication February 2009.

Accepted for publication September 2009.

Address for correspondence: Irene Kane, Ph.D., University of Pittsburgh School of Nursing, 350 Victoria Building 422, 3500 Victoria Street, Pittsburgh, PA 15261; E-mail:

Article Outline

Epidemiological data for the period 1991-2003 indicate that the decrease in physical activity observed in American children is a major contributor to the unprecedented rise in obesity for this age group (27,32). Stephens (27) has issued a call to action to reverse the physical inactivity crisis of American children. Understanding factors that determine children's physical activity behavior has been deemed critical to promoting physical activity among American youth (3,25,26,30).

An initial step in developing long-term intervention strategies to initiate and to maintain children's physical activity participation requires identification of underlying psychological constructs that shape movement behavior (12,26,28,29). It is proposed that one such psychological construct is exercise discomfort. Buckworth and Dishman (7) observed that exercise-related discomfort was negatively correlated with self-report of physical activity level. Therefore, the present investigation examined exercise discomfort in middle school children performing a standardized aerobic field test, the Progressive Aerobic Cardiovascular Endurance Run (PACER) shuttle run. The choice of middle school students as research subjects was based on previous research (1,11,32), indicating that physical activity decreases during adolescence and may in turn be associated with increases in body weight disproportionate to growth status. As such, it was considered important to examine the psychological construct of discomfort that may operate at the onset of this developmental period to influence physical activity behavior (25,28,30,32).

In the context of this investigation, exercise discomfort is defined by the construct-specific measurements of perceived exertion and muscle pain. Exertional perception has been described as a complex sensory process influenced by physiological, psychological, cognitive, and social or situational mediators (22,23). Comparatively, more intense perceptions of exertion can negatively influence physical activity adoption and adherence (23). Neuromuscular pain has generally been described as an unpleasant sensation associated with actual or potential tissue damage (23). The naturally occurring skeletal muscle pain that occurs during exercise is associated with increasing aerobic or resistance exercise intensity (6,9,16,23,24). Cook et al. (9) and O'Connor and Cook (16) concluded that exercise-induced naturally occurring muscle pain is a critical factor in determining exercise performance. It should be noted that the word "hurt" was the descriptor of choice in the present investigation to define exercise-induced skeletal muscle pain because it is commonly understood by children as young as 3 yr (36).

The Exercise Discomfort Index (EDI) for children was calculated as a rating of perceived exertion for the overall body (RPE-O) multiplied by a rating of muscle hurt (RMH). The rationale for using the product of these scores was to quantify exercise discomfort with appropriate metric attributes resulting in a commonly used 0-100 scale easily interpreted by children (36).

Poulton et al. (18) used a cognitive appraisal model on the basis of a match-mismatch paradigm to compare predictions of exercise discomfort with actual exercise discomfort reports in young adults. Discomfort was measured with a single question, that is, "how much discomfort do you anticipate experiencing/did you experience?" Discomfort predictions were considered correct when they matched the actual reports. Incorrect predictions were identified by mismatches, that is, overprediction or underprediction relative to the actual exercise report. In their longitudinal study, patterns of overprediction of exercise discomfort were related to lower levels of physical activity in young adults (18). The young adult subjects avoided exercise on the basis of cognitions that formed perceptual predictions of undesirable discomfort. Poulton et al. (18) concluded that physical inactivity may be associated with specific cognitions, such as a mismatch between predicted and actual exercise-induced discomfort. It was observed that young adults who were overpredictors generally had lower levels of physical activity and unfavorable physical health measures, for example, excess weight and body mass index (BMI) (18). In the present study, a match-mismatch paradigm was used to compare children's predicted exercise discomfort with that actually perceived during exercise. Physical activity history and BMI data were used as general assessments of subjects' health-related fitness. The relation between these health fitness measures and exercise discomfort predictions was subsequently determined.

Insight into children's predictions of exercise discomfort may contribute to the development of interventional strategies that promote children's physical activity participation and deter a pattern of exercise avoidance with its concomitant loss of physical and mental health benefits. As such, it was hypothesized that 1) children's predicted aerobic exercise discomfort would be greater than actual aerobic exercise discomfort and 2) discomfort predictions would be positively related to BMI and negatively related to physical activity and sport participation. It was expected that these findings would be observed for both female and male children.

Back to Top | Article Outline



The sample was comprised of 34 healthy female (n = 18) and male (n = 16) children ages 11-14 yr who volunteered as subjects in this investigation. Descriptive characteristics including BMI, physical activity, and sport participation are summarized in Tables 1 and 2. Subjects were recruited from the Falk Laboratory School affiliated with the University of Pittsburgh (grades 6-8). Clinical, neuromotor, or cognitive contraindications to exercise testing were not reported by the school nurse, child, or parents nor observed by the investigators. No subject who volunteered was declined participation. Written informed consent to participate was obtained from each subject and their respective parent or guardian. All experimental procedures and related participation consent documents complied with the Falk Laboratory School Policy and Procedure (33) and the Human Use Guidelines of the American College of Sports Medicine (2) and were approved by the University of Pittsburgh's institutional review board (33).

Back to Top | Article Outline
Experimental Design

This investigation used a within-group pretest and posttest experimental design. Pretest and posttest were administered on separate days scheduled 48 h apart. Each test session required 45 min and was administered in the same gymnasium.

Back to Top | Article Outline
Testing Procedures

During the pretest session on day 1, subjects were oriented to the PACER, which included 1) listening to a CD containing a description of the PACER, 2) PACER starting instructions, and 3) music that set the pace of the PACER test. Subjects then viewed a PACER demonstration video. Immediately after the PACER orientation, subjects reported their predicted EDI. Each subject then practiced the PACER as recommended in the FITNESSGRAM® Administration Manual (15). On day 2, the PACER exercise test was administered, followed immediately by the posttest session to assess actual EDI.

Back to Top | Article Outline
Exercise Discomfort Index (EDI)

The product of RPE-O and RMH was calculated to determine the EDI for the predicted and actual discomfort measurements, that is, EDI = RPE-O × RMH.

Back to Top | Article Outline
Rating of perceived exertion.

The Children's OMNI Scale of Perceived Exertion, walk/run format (Fig. 1), was used to assess an undifferentiated RPE for the overall body (RPE-O) (23). This rating was reported as a whole number from 0 to 10, with 0 indicating no exertion and 10 indicating maximal exertion for the overall body. Ratings were obtained using standardized definitions, instructions, and anchor procedures (23). The predicted RPE-O was obtained during the pretest session. This assessment occurred immediately after presentation of the PACER instructions and the PACER video but before the PACER practice session. Upon completion of standardized RPE-O orientation procedures, subjects were instructed, "Based on the description of the PACER you have received, I would like you to predict how tired you think you might feel when running the PACER." On day 2, actual RPE-O was measured during the posttest session immediately after completion of the PACER test. After presentation of standardized RPE-O procedures, subjects were instructed, "Please select the number that tells how tired your overall body felt when running the PACER."

Back to Top | Article Outline
Rating of muscle hurt.

Subjects were requested to rate their perception of leg muscle hurt (RMH) using the OMNI Scale of Muscle Hurt (Fig. 2) (24). The RMH was reported as a whole number from 0 to 10, with 0 indicating no hurt and 10 indicating maximal leg muscle hurt (RMH). Standardized instructions specific to walk/run exercise and to the subject's age-related comprehension included definition and anchoring procedures adapted from directions previously developed for the Children's OMNI Scale of Muscle Hurt during resistance exercise (24). The predicted RMH was obtained during the pretest session immediately after viewing the PACER video and determination of RPE-O. After presentation of standardized RMH procedures, subjects were instructed, "Based on the description of the PACER you have received, please predict how much hurt you think you may feel in your leg muscles while running the PACER." On day 2, actual RMH was measured during the posttest session immediately after completion of the PACER test. Again, after presentation of standardized RMH procedures, subjects were instructed, "I am defining muscle hurt as the amount or intensity of hurt that you did feel in your leg muscles during the PACER exercise. Please use the picture before you to describe how much your leg muscles hurt during the PACER shuttle run." The RMH always followed the RPE.

Back to Top | Article Outline
PACER Exercise Test

The PACER shuttle run is a 20-m graded fitness test administered using CD-recorded music and auditory beeps that present 21 levels (21 min) of progressively more intense running (15). Each PACER running level requires 1 min for completion. The 20-m PACER shuttle run, validated for boys and girls aged 8-19 yr, served as the exercise forcing function for the measure of perceived discomfort (4,14).

The PACER exercise test was administered at the beginning of the second day. The total subject pool was initially divided into testing groups of seven to eight individuals using random assignment. The first randomly selected group of seven to eight subjects was then instructed to line up on the prepared course. When the first group was in place, they were instructed to listen carefully to the instructional CD and be ready to run as soon as they heard the CD speaker state "on your mark…get ready…'START.'" Subjects ran back and forth across a 20-m distance within a defined lane to an increasing pace set to music and beeps (15). Subjects were instructed that the test was concluded the second time that he or she failed to reach one of the two end lines within their lane before the beep sounded (15). As soon as the subject completed his or her last lap, the actual EDI was immediately obtained and recorded by the assigned investigator.

The score for the PACER exercise test is the number of completed laps (Table 2). The completed lap count was determined by an investigator positioned behind the start or end line and assigned to record the performance of a specific subject. The count was recorded using the standardized FITNESSGRAM® lap scoring sheet (15).

Back to Top | Article Outline
Data Analysis

All statistical analyses were conducted using SPSS for Windows (version 14.0; SPSS, Inc., Chicago, IL). A two-way (gender × assessment time point) ANOVA with repeated measures on assessment time point was used to examine the difference between predicted and actual EDI and to determine whether the magnitude of the difference varied by gender. Height, weight, BMI, and completed PACER laps were calculated as means ± SD. Data obtained from the Physical Activity and Sport Participation Questionnaire were summarized using medians as recommended (1).

Pearson correlation coefficients were used to examine relations between selected variables. A Bland-Altman plot assessed the level of agreement between the predicted and the actual EDI and displayed the differences between the predicted and the actual EDI for individual subjects. BMI and PACER laps for subjects who either overpredicted or underpredicted exercise discomfort were subsequently compared using independent sample t-tests. Physical activity and sport participation level for subjects who either overpredicted or underpredicted exercise discomfort were compared using chi-square and Mann-Whitney U tests. The level of significance was set at 0.05 (two-tailed).

Power analysis was calculated using the sample size tables by Cohen with a one-tailed alpha of 0.05 and an estimated effect size of 0.7 and a desired power of 0.90 (8). The estimate of effect size was based in part on a previous study of perceived exertion in children aged 10-14 yr, which found a mean actual RPE-O score of 5.6 with an SD of 2.0 (34). Because it was anticipated that children in general would overpredict RPE-O, a mean predicted RPE-O of 7.0 was posited, leading to an estimated effect size of 0.7 (7.0-5.6)/ 2 = 7). The total sample size required to detect an effect size of 0.7 with a power of 0.90 was 36 subjects.

Back to Top | Article Outline


Back to Top | Article Outline
Predicted and actual exercise discomfort.

Descriptive statistics for predicted and actual EDI data are reported in Table 3. The ANOVA revealed a significant main effect for assessment time point (F1,32 = 5.9, P = 0.021, η2 = 0.16), indicating that the predicted EDI was significantly higher than the actual EDI. The main effect for gender was not significant (F1,32 = 3.52, P = 0.070). The interaction effect between gender and assessment time point was also not significant (F1,32 = 0.19, P = 0.668).

A Bland-Altman plot was constructed to examine limits of agreement between the predicted and the actual EDI (Fig. 3). The mean difference (or bias) is indicated by a solid line. The 95% confidence interval (±1.96 SD; upper and lower limits of agreement) is designated by dashed lines. The comparatively large width of the confidence interval indicated poor agreement between predicted and actual EDI when examined on an individual basis for the mixed gender sample. It can also be seen in Figure 4 that variability increased as the mean predicted and actual EDI increased.

The main effect of assessment time point on RPE-O was not significant (F1,32 = 0.12, P = 0.729). The interaction effect between gender and assessment time point was also not significant (F1,32 = 3.53, P = 0.069). However, the main effect of gender was significant (F1,32 = 5.26, P = 0.029, η2 = 0.14). As seen in Table 4, females overpredicted RPE-O whereas males underpredicted RPE-O. The main effect of assessment time point on RMH was significant (F1,32 = 9.43, P = 0.004, η2 = 0.23). However, neither the main effect of gender (F1,32 = 1.63, P = 0.221) nor the interaction effect between gender and assessment time point (F1,32 = 0.04, P = 0.837) was significant. It can be seen in Table 4 that both males and females significantly overpredicted RMH.

A strong positive correlation (P < 0.001) was found between the predicted and the actual EDI for the total (r = 0.701, P = 0.000), female (r = 0.699, P = 0.001), and male (r = 0.660, P = 0.003) groups. These data indicated that subjects who expected more discomfort reported experiencing more discomfort during the PACER exercise.

Back to Top | Article Outline
Discomfort predictions and demographic characteristics.

Idiographic analysis allowed the identification of those subjects who either overpredicted or underpredicted their EDI compared with their actual EDI. Underprediction of discomfort was defined as a lower predicted EDI than the actual EDI. Overprediction of discomfort was defined as a higher predicted EDI than the actual EDI. Figure 4 displays the distribution of individuals who overpredicted and underpredicted discomfort. The diagonal line represents equality between predicted and actual values. The three data points that fall exactly on the line represent the three subjects (one female and two males) whose predicted and actual EDI were the same. Data points that fall above the diagonal line represent the 23 subjects (14 females and 9 males) who overpredicted discomfort, whereas data points that fall below the diagonal line represent 8 subjects (3 females and 5 males) who underpredicted discomfort.

Those subjects who either overpredicted or underpredicted exercise discomfort were in turn compared according to BMI, PACER laps completed, and physical activity and sport participation level. Presented in Table 5 are data indicating whether subjects who overpredicted discomfort differed from subjects who underpredicted discomfort with respect to these selected demographic variables. A significant difference between overpredictors and underpredictors was found for only one variable, hours of leisure physical activity per week (P = 0.038). Subjects who underpredicted exercise discomfort reported a median of 11.14 h·wk−1 of recreational activity. Those subjects overpredicting exercise discomfort reported a median of 5.25 h·wk−1 of recreational activity. Neither BMI, PACER laps completed, nor other physical activity measures differed between overpredictors and underpredictors of exercise discomfort.

Back to Top | Article Outline


Back to Top | Article Outline
Predicted and actual exercise discomfort.

This study used a match-mismatch paradigm to examine children's predicted and actual exercise discomfort associated with performance of an aerobic physical activity of progressively increasing intensity. The match-mismatch paradigm involved a cognitive appraisal process that compared predicted perceptions to actual perceptions measured, respectively, before and after a specific physical activity, that is, the PACER (18). Mean data for the total group indicated an overprediction (P = 0.021) of exercise discomfort associated with performance of the PACER. That is, the predicted EDI was higher than the actual EDI. This finding was consistent with the research hypothesis that the children's anticipated level of discomfort during shuttle run exercise would be greater than the actual discomfort that they perceived and reported. Because the present perceptual responses were very similar to the group-normalized RPE-O of 6 (OMNI Scale) that generally corresponds to the anaerobic threshold as measured by the ventilatory breakpoint (23), the exercise intensity of the PACER was presumably adequate to produce exercise-related discomfort that the children accurately anticipated.

The upper and the lower limits of agreement (i.e., 95% confidence interval) for predicted and actual EDI derived from the Bland-Altman plot for the total group were 35.68 and 22.92, respectively (Fig. 3). It would be expected that 95% of the computed difference values for discomfort (i.e., predicted minus actual) would fall between these limits. The comparatively large width of the confidence intervals indicated poor agreement between predicted and actual discomfort for the total group. These results provided further evidence that subjects predicted a greater level of discomfort than what was actually experienced when performing the submaximal PACER shuttle run.

Using a match-mismatch paradigm to examine predicted and actual exercise discomfort of middle school children performing the PACER shuttle run was consistent with paradigms used in previous research. The findings observed presently for 11- to 14-yr-old children were generally similar to those reported by Poulton et al. (18) for young adults. Interestingly, Poulton et al. (18) reported that young adult females overpredicted exercise discomfort, whereas young adult males underpredicted exercise discomfort. However, Poulton et al. (18) used the term "undefined" discomfort in conjunction with exercise participation. The current study used a more specific measure by operationally defining exercise discomfort as the product of an RPE for the overall body and a rating of leg muscle hurt. It was expected that the EDI calculated as the product of predicted and actual RPE-O and RMH would provide a more task-specific measurement of exercise discomfort in determining match-mismatch responsiveness.

Back to Top | Article Outline
RPE-O and RMH responses.

Statistical analysis for the main effect of assessment time point demonstrated that predicted (5.96 ± 2) and actual (6.03 ± 2) RPE-O did not differ, indicating a response match. Subjects in the current investigation reported at least moderate experience with various types of aerobic exercise. Such participation may have resulted in an accurate expectation of exertion that was based on their repeated exposure to running as part of their normal daily recreational and sport activities (1,23).

The statistical analysis determined a significant gender effect for RPE-O. This finding indicated that the female subjects rated their perceptions of exertion higher than the male subjects at both the predicted and the actual assessment points. The gender difference in RPE-O may have been linked to a comparatively lower aerobic fitness level for the female than male children (23). In the present investigation, the number of PACER laps completed served as a surrogate measure of aerobic fitness, with males completing 29.8 laps and the females completing 21.4 laps (Table 2). It was speculated that the females had a lower level of aerobic fitness on the basis of the comparatively fewer number of laps completed. The female children's lower level of aerobic fitness may have resulted in higher perceptions of exertion at any given time point during the shuttle run. In a study of 9- to 17-yr-old children performing incremental treadmill exercise, Robbins et al. (21) noted that females tended to report greater exertion than males during each of the study periods. They attributed this response to lower fitness levels in the females (21). Interestingly, Robertson also noted that when adult males and females performed at the same submaximal exercise intensity, RPE-O was higher for the individuals who had the lower aerobic fitness level (23).

In contrast, RMH in the present study was significantly overpredicted. The group mean predicted RMH was greater than actual RMH (predicted = 3.98 ± 2, actual = 2.93 ± 2, P = 0.004). Muscle hurt was significantly overpredicted, resulting in a response mismatch. This response was consistent with the clinical observation of Rachman and Arntz (19) who reported that individuals generally overpredict the pain they expect to experience. However, Rachman and Lopatka (20) emphasized that accurate sensory predictions must be learned. Accurate prediction of pain is important in preventing avoidance behavior regarding painful events, possibly involving aerobic exercise of the type used in the present investigation (20).

Previous investigations emphasized the need to develop strategies to assist children in making accurate pain predictions (13). Children have been shown to evidence good recall for experiences (e.g., weather, illness, and physical exertion) associated with painful events (35). In an analogous manner, it is critical that children receive accurate and easily understood information before initiating exercise to make appropriate predictions regarding the level of discomfort to be experienced during physical activity participation (13,35,36). Such preparation can result in a match between favorable levels of predicted and actual sensory exercise experience. This in turn could prevent avoidance of exercise experiences on the basis of overprediction of expected pain sensations (17,18,35). Therefore, it is suggested that cognitive and/or behavioral interventions be developed to help children learn to expect and to accept moderate and tolerable levels of exercise discomfort to promote rather than avoid exercise participation.

In a similar context, overprediction of discomfort may lead to avoidance or lower levels of exercise participation (17-19). Physical activity avoidance by both adults and children can be influenced by beliefs and memories of discomfort experiences. This may lead to an overall preference for reduced discomfort (17). For example, the correlation between predicted and actual EDI observed presently for the total group was consistent with the perceptual predictions of anticipated discomfort observed by Poulton et al. (18). In their investigation involving young adults, predictions of anticipated discomfort were subsequently linked to exercise avoidance.

Combining RPE-O and RMH in the current study to create an EDI was intended to provide a more robust measure of the potentially aversive sensory milieu during aerobic exercise (23). Personal knowledge of their EDI could assist children in forming a more accurate "prediction adjustment" regarding an ensuing aerobic exercise experience (10,18,35). The present findings are consistent with previous research that stated exercise-induced pain and exertional perception are distinct sensory domains that occur more or less simultaneously during exercise (23). In the present investigation, the middle school female and male children a) accurately predicted their overall body exertion, although the comparatively more intense perceptual ratings by females suggest the need for further research, and b) overpredicted muscle hurt while performing the PACER shuttle run. This mismatch is consistent with the observation of O'Connor and Cook (16) that naturally occurring leg muscle pain at peak exercise elicited greater cognitive-emotional response when compared with other noxious stimuli, for example, exertion and heat or cold stimulation. From these findings, it is proposed that the perception of muscle hurt was the primary factor in shaping the predicted level of discomfort for aerobic exercise. The naturally occurring skeletal muscle hurt increased with increasing exercise intensity (6,9,23). The overprediction of muscle hurt observed presently occurred in the presence of accurate predictions of exertional perceptions (6,9,24). The comparatively greater muscle hurt resulted in an overprediction of expected discomfort during performance of the 20-m shuttle run.

On the basis of the present findings, it is proposed that 1) the EDI can be used to objectively assess a previously undefined but widely used perceptual description of exercise-related sensations, that is, discomfort; and 2) exertion and muscle hurt are parallel but not isomorphic perceptual constructs that influence children's discomfort perceptions regarding aerobic exercise. Specifically, it was hypothesized that exercise discomfort operated as a psychological construct that could potentially influence middle school children's willingness and motivation to perform a common aerobic activity. In general, the findings reinforced the research of Sallis et al. (25,26) that psychological constructs such as activity-specific beliefs, for example, perceptions of exertion, may be strongly associated with or predictive of children's level of physical activity participation. Finally, the results supported the observation of Poulton et al. (18) that using a match-mismatch paradigm involving predicted and actual sensory measurements may be useful for investigation of factors associated with the initiation of exercise (25,29).

Back to Top | Article Outline
Discomfort predictions and selected demographics

Poulton et al. (18) noted that young adults who were overpredictors of expected discomfort reported fewer days on which they engaged in at least 30 min of physical activity and had more negative scores on physical health measures (e.g., weight, BMI) (18). The present investigation examined differences in BMI, physical activity and sport participation, and PACER laps completed between subjects who underpredicted and overpredicted exercise discomfort. The data indicated that the subjects in the present study were within acceptable criterion-referenced physical fitness standards (2,5,31). Subjects did not show comparatively more negative scores on physical health measures as noted in previous research. Because this study may have attracted a selected sample of physically active middle school students who were of normal body weight based on national standards (31), selection bias could have accounted for the absence of a discomfort prediction relation with BMI. However, results indicated one statistically significant finding: subjects who underpredicted discomfort relative to actual exercise reports participated in a greater number of hours per week of recreational activity than those who overpredicted discomfort.

Therefore, the present results regarding perceived discomfort were in agreement with the observation of Poulton et al. (18) that specific physical activity-related cognitions may predispose physical inactivity. Overpredictors in the present study reported lower levels of recreational physical activity participation. This is indicative of the physical activity avoidance behavior that has been shown to occur in individuals who overpredict expected exercise discomfort (18-20).

Back to Top | Article Outline
Conclusions and recommendations.

The sample of middle school children in this study predicted greater aerobic exercise discomfort than they actually experienced when performing a PACER test. This finding suggests that understanding psychological processes that are linked to aerobic physical activity participation of children is critical to ensuring an accurate applied psychology knowledge base for successful long-term interventional strategies (12). Examining exercise discomfort with a match-mismatch paradigm followed the suggestion of Poulton et al. (18) to study specific thought processes and attitudes in close temporal proximity to physical activity to understand factors that lead to the initiation and maintenance of a physically active lifestyle. No one variable can be expected to account for a child's physical activity behavior. However, knowledge about a child's predicted and actual perceptions of exercise discomfort provides a valuable first step in understanding the role that discomfort perceptions play in shaping children's beliefs about participation in aerobic physical activity. It is possible that a discomfort construct plays an important role in the initiation and maintenance of aerobic exercise by middle school children. Such findings can in turn inform physical activity interventions and/or innovative health fitness components of physical education curricula intended to promote cardiovascular health and fitness through regular participation in aerobic physical activity.

Several limitations of the present study should be noted: 1) use of a convenience sample may have attracted middle school students who were favorably predisposed to exercise participation, 2) sample size within each gender limits generalization of findings, 3) aerobic fitness level was estimated and not directly determined, 4) aerobic physical activity status was obtained from a physical activity and sport participation self-report, 5) possible effects of social referencing and group conformity during the PACER may have influenced subject performance, and 6) a less controlled field test setting as contrasted to a controlled laboratory setting was used.

It is recommended that future research use a match-mismatch paradigm to study exercise discomfort using different exercise modes, subject demographics, larger sample by gender, and individual testing of subjects. Further, exploration of the effect of an intervention strategy using the EDI is encouraged. Providing children with a cognitive reference upon which to accurately assess their actual aerobic exercise discomfort and monitoring exercise discomfort changes as a function of a conditioning program may contribute to improving physical activity-related perceptions which encourage exercise participation.

This study was funded by a grant from the University of Pittsburgh School of Education.

The authors thank Ms. Laura Hunt, Falk physical education teacher, for her assistance.

The results of the present study do not constitute endorsement by the American College of Sports Medicine.

Back to Top | Article Outline


1. Aaron DJ, Storti KL, Robertson RJ, Kriska AM, LaPorte RE. Longitudinal study of the number and choice of leisure time physical activities from mid to late adolescence. Arch Pediatr Adolesc Med. 2002;156(11):1075-80.
2. American College of Sports Medicine. Guidelines for Exercise Testing and Prescription. 6th ed. Philadelphia (PA): Lippincott Williams & Wilkins; 2000. p. 214-23.
3. Barnett TA, O'Loughlin J, Paradis G. One- and two-year predictors of decline in physical activity among inner-city schoolchildren. Am J Prev Med. 2002;23(2):121-7.
4. Baumgartner TA, Jackson AS, Mahar MT, Rowe DA. Measurement for Evaluation in Physical Education and Exercise Science. Boston (MA): McGraw Hill; 2007. p. 340-8.
5. Biddle S, Sallis J, Cavill N. Young and active? Young people and health-enhancing physical activity-evidence and implications. In: A Report of the Health Education Authority Symposium Young and Active. London (England): Health Education Authority; 1998. p. 1-16.
6. Borg G, Ljunggren G, Ceci R. The increase of perceived exertion, aches, and pain in the legs, heart rate and blood lactate during exercise on a bicycle ergometer. Eur J Appl Physiol. 1985;54(4):343-9.
7. Buckworth J, Dishman RK. Exercise Psychology. Champaign (IL): Human Kinetics; 2002. p. 200-2, 218-9.
8. Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale (NJ): Lawrence Erlbaum Associates; 1988. p. 39-54.
9. Cook DB, O'Connor PJ, Eubanks SA, Smith JC, Lee M. Naturally occurring muscle pain during exercise: assessment and experimental evidence. Med Sci Sports Exerc. 1997;29(8):999-1012.
10. Dannecker EA, Price DD, Robinson ME. An examination of the relationships among recalled, expected, and actual intensity and unpleasantness of delayed onset muscle pain. J Pain. 2003;4(2):74-81.
11. De Bourdeaudhuij I, Lefevere J, Deforche B, Wijndaele K, Matton L, Philippaerts R. Physical activity and psychosocial correlates in normal weight and overweight 11-19 year olds. Obes Res. 2005;13(6):1097-105.
12. Kimiecik JC, Blissmer B. Applied exercise psychology: measurement issues. In: Duda JL, editor. Advances in Sport and Exercise Psychology Measurement. Morgantown: Fitness Information Technology, Inc; 1998. p. 447-60.
13. Lander J, Hodgins M, Fowler-Kerry S. Children's pain predictions and memories. Behav Res Ther. 1992;30(2):117-24.
14. Liu N, Plowman S, Looney M. The reliability and validity of the 20-meter shuttle test in American students 12 to 15 years old. Res Q Exerc Sport. 1992;63(4):360-5.
15. Meredith MD, Welk GJ. Aerobic capacity: the PACER. In: Meredith MD, Welk GJ, editors. FitnessGram, ActivityGram: Test Administration Manual. Champaign (IL): Human Kinetics; 2005. p. 27-32.
16. O'Connor PJ, Cook DB. Exercise and pain: the neurobiology, measurement, and laboratory study of pain in relation to exercise in humans. Exerc Sport Sci Rev. 1999;27(1):119-66.
17. Philips HC. Avoidance behaviour and its role in sustaining chronic pain. Behav Res Ther. 1987;25(4):273-9.
18. Poulton R, Trevena J, Reeder AI, Richard R. Physical health correlates of overprediction of physical discomfort during exercise. Behav Res Ther. 2002;40(4):401-14.
19. Rachman S, Arntz A. The overprediction and underprediction of pain. Clin Psychol Rev. 1991;11(4):339-55.
20. Rachman S, Lopatka C. Accurate and inaccurate predictions of pain. Behav Res Ther. 1988;26(4):291-6.
21. Robbins LB, Pender NJ, Ronis DL, Kazanis AS, Pis MB. Physical activity, self-efficacy, and perceived exertion among adolescents. Res Nurs Health. 2004;27(6):435-46.
22. Robertson RJ, Noble BJ. Perception of physical exertion: methods, mediators, and applications. Exerc Sport Sci Rev. 1997;25(1):407-52.
23. Robertson RJ. Perceived Exertion for Practitioners-Rating Effort with the OMNI Picture System. Champaign (IL): Human Kinetics; 2004, p. 2-18, 101, 107-8, 136-8, 146.
24. Robertson RJ, Goss FL, Aaron DJ, et al. Concurrent muscle hurt and perceived exertion of children during resistance exercise. Med Sci Sports Exerc. 2009;41(5):1146-54.
25. Sallis JF, Prochaska JJ, Taylor WC. A review of correlates of physical activity of children and adolescents. Med Sci Sports Exerc. 2000;32(5):963-72.
26. Sallis JF, Simons-Morton BG, Stone EJ, et al. Determinants of physical activity and interventions in youth. Med Sci Sports Exerc. 1992;24(6):S248-57.
27. Stephens M. Children, physical activity, and public health: another call to action. Am Fam Physician. 2002;6(6):1033-5.
28. Stone EJ, McKenzie TL, Welk GJ, Booth ML. Effects of physical activity interventions in youth: review and synthesis. Am J Prev Med. 1998;15(4):298-315.
29. Trost SG, Pate RR, Ward DS, Saunders R, Riner W. Correlates of objectively measured physical activity in preadolescent youth. Am J Prev Med. 1999;17(2):120-6.
30. Trudeau F, Laurencelle L, Shephard R. Tracking of physical activity from childhood to adulthood. Med Sci Sports Exerc. 2004;36(11):1937-43.
31. U.S. Department of Health and Human Services Centers for Disease Control and Prevention Web site [Internet]. Atlanta (GA): CDC; [cited 2007 July]. Available from:
32. U.S. Department of Health and Human Services Centers for Disease Control and Prevention. Youth risk behavior surveillance-United States, 2003. May 21, 2004. MMWR Morb Mortal Wkly Rep. 2004:53(SS2):1-30.
33. University of Pittsburgh Web site [Internet]. Pittsburgh (PA): Falk Laboratory School; [cited 2007 Jan 6]. Available from:∼fls/; Institutional Review Board Guidelines; [cited 2007 Jan 6]. Available from:
34. Utter AC, Robertson RJ, Nieman DC, Kang J. Children's OMNI Scale of Perceived Exertion: walking/running evaluation. Med Sci Sports Exerc. 2002;34(1):139-44.
35. Von Baeyer CL, Marche TA, Rocha EM, Salmon K. Children's memory for pain: overview and implications for practice. J Pain. 2004;5(5):241-9.
36. Wong DL, Hockenberry-Eaton M, Wilson D, Winkelstein ML. Wong's Essentials of Pediatric Nursing. 6th ed. St. Louis (MO): Mosby, Inc; 2001. p. 1046-69.


©2010The American College of Sports Medicine