psychological affect at different ratings of perceived exertion in high

Perceptzral and Motor Skills, 1996, 82, 1035-1042

O l'erceptual and Motor Skills 1996

PSYCHOLOGICAL AFFECT AT DIFFERENT RATINGS OF PERCEIVED EXERTION IN HIGH- AND LOW-ACTIVE WOMEN: A STUDY USING A PRODUCTION PROTOCOL ' GAYNOR PARFITT, ROGER ESTON, AND DECLAN CONNOLLY

Division of Health and Human Perforrnarrce Uniuerrit~of Wales Sum~nary.-The purpose of this study was to examine psychological affect at .different ratings of perceived exertion ( W E ) in 15 high- and 15 low-active women. Both groups performed three steady-state exercise bouts on a cycle ergometer at W E s 9, 13. and 17 and r e ~ o r t e dtheir affect in the last 20 sec. of and 5 min. after each work rate. There were no differences between groups in percentage of maximal oxygen uptake ( % V02max) at each W E . Low-accive women reported hehg significantly more negative at W E 17 than W E 9 and less positive than the high-acuve women at W E s 9, 13, and 17. In addition, all subjects reported more positive feelings 5 rnin. postexercise than in the last 20 sec. of exercising, especially at W E 17. These results have implications for exercise prescription in groups ddfering in habitual activity levels.

A major challenge to exercise leaders and health promoters is to try to encourage low-active indviduals to become regular exercisers. This paper extends previous work by Parfitt and colleagues (Parfitt, Markland, & Holmes, 1994; Parfitt & Eston, 1995a; Parfitt & Eston, 1995b) which considered the psychological affect of high- and low-active subjects at different intensities of exercise. In the first two studes, intensity of exercise was regulated using a submaximal aerobic power test, Astrand Cycle Ergometer Test (Astrand, 1988), while in the latter ratings of perceived exertion ( W E ) were used in a production protocol. Psychological affect was measured in a1 three studies using Rejeslu, Best, Griffith, and Kenney's (1987) F e e h g Scale. In their results, Parfitt, et al. (1994) and Parfitt and Eston (1995a) indicated that highand low-active subjects differ in psychological affect when working at moderate and high intensities of exercise. The high-active subjects (both men and women) were significantly more positive in the 90% work-rate condition than the low-active subjects. However, there was no difference between the groups in the 60% work-rate condition ( p < ,001). The affect or dzstress schema in Leventhal and Everhart's (1979) parallel-processing model was proposed as one explanation for the group differences at the higher work rate, with low-active subjects having a more negative affective schema concerning exercise at high intensities. Psychological affect was also assessed at ddferent temporal points in the 'Address correspondence to Dr. Ga nor Parfitt, Division of Health and Human Performance, University of Wales, Bangor, ~ f r i d d b e d dBuilding, Victoria Drive, Bangor, Gwynedd, LL57 2EN, Wales.

1036

G. PARFITT. ET AL.

exercise sessions: during and after exercising at each work rate (Parfitt, et

al., 1994) and after 2 min. and 4 min. at each work rate (Parfitt & Eston, 1995a). Psychological affect was significantly lower in the last minute of exercising at the 90% work-rate condition compared to any other time or work rate. This temporal factor may also be related to the affect or &stress schema, with the physiological cues being more acute during the exercise bout but decreasing after cessation of exercise. Importantly, this implies that, regardless of experience, the 90% work rate was interpreted negatively during the exercise bout but relatively positively afterwards. It could be argued that these studies were limited insofar as only two work rates were considered. With a greater range of exercise work rates the differences between the high- and low-active individuals may become more marked. The authors have already recognised the hitations of the Wstrand Cycle Ergometer Test (Astrand, 1988) for controhng intensity of exercise, measured as percentage of maximal oxygen uptake (Parfitt & Eston, 1995a). Further, from a perspective of exercise promotion, it may be useful to consider these relationships when the subject independently sets the intensity regulated by the W E rather than having the work rate adjusted by the investigator and the subject being requested to report the W E at each stage. Consequently, Parfitt and Eston (1995b) conducted a pilot study in which high- and low-.active women were required to exercise using a production protocol based upon three standardized W E s (9, 13, and 17). In healthy subjects, several studies have confirmed the validity of self-regulation of intensity of exercise guided by the 15-point RPE scale (Borg, 1970) for treadmill running (Eston, Davies, & Wihams, 1987; Dunbar, Robertson, Baun, Blandin, Metz, & Burdett, 19921, and cycling (Eston & W&ams, 1988; Dunbar, et al., 1992). Parfitt and Eston's (1995b) results confirmed that exercise intensities of RPEs 9 and 13 in low-active subjects elicited the most positive psychological affect. These results sit comfortably with previous research concerned with intensity of exercise and emotional response (e.g., Parfitt, et a/., 1994; Moses, Steptoe, Mathews, & Edwards, 1989). However, in this pilot study the increased range of exercise intensities d d not result in a change in the pattern of results from moderate to high exercise intensities in the lowand high-active subjects. It appeared that the low intensity of exercise (WE 9) was associated with the same positive affect as the moderate intensity ( W E 13) for both groups. This could be attributed to the fact that the lowactive individuals were working at different relative exercise intensities despite similar WEs. This has been observed recently by Dishman, Farquhar, and Cureton (1994). Cross-sectional studies have also shown that trained subjects report lower W E s at a given percentage of maximal oxygen uptake than untrained subjects (DemeUo, Cureton, Boineau, & Singh, 1987; Boutcher, Seip, Hetzler, Pierce, Snead, & Weltman, 1989). The purpose of this

AFFECT, PERCEIVED EXERTION

1037

study was to clarlfy this by measuring oxygen uptake and the percentage of maximal oxygen uptake at the different W E s in low- and high-active women. It was hypothesised that low-active subjects would work at a lower proportion of the maximal oxygen uptake for a glven W E and report more with greater exertion. negative affect than the high-active subjects, espec~all~ It was also hypothesised that affect would be more negative in the last 20 sec. of exercise than 5 min. postexercise. Subjects Subjects (N=3O) were self-reported high-active ( n = 15) and low-active (n=15) women. High-active subjects were those who exercised three or more times for more than 20 min. per week. Low-active subjects were those who exercised once a week or less. Table 1 (cf. p. 1039) provides biometric data on the subject population. None of the subjects were regular bicycle riders. Measures Self-report questionnaire.-This was used to measure physical activity. Subjects indicated how frequently and how long they participated in physical activity each week and whether this had been typical behavior over recent years. Subjects were informed that activities such as running, cychg, swimming, aerobics, squash and team games constituted physical activity in this instance. Subjects also provided details of their date of birth and contact address. Rating of perceived exertion.-General, i.e., total inner f e e h g of effort, ratings of perceived exertion ( W E ) were assessed with the Borg 6-20 Category Scale (Borg, 1970). The WE was used as the independent variable to regulate work rates corresponding to W E s of 9, 13, and 17. Feeling Scale.-Rejesh and colleagues' (Hardy & Rejesh, 1989; Rejeski, et al., 1987) Feelmg Scale, which measures psychological affect during exercise, was used. This is a bipolar scale that ranges from + 5 to -5 with verbal anchors of + 5 =very good, + 3 = good, + 1 = slightly good, O =neutral, -1 = slightly bad, -3 =bad, and -5 =very bad. The instructions followed those of Hardy and Rejeski (1989, p. 309) with the addition of one phrase. Following the sentence, "Additionally, feehgs may fluctuate across time," the phrase "and after" was added to the next sentence so that it read: "That is, one might feel good and bad a number of times during and after exercise." This phrase was added to enable changes in psychological affect to be assessed both during and after the exercise session. Procedure.-The experiment was conducted incGvidually in the exercise

1038

G. PARFlTT, ET AL.

physiology laboratory. A four-phase protocol was used in the data collection: Famiharisation, Trial 1, Trial 2, and Trial 3. Each of these phases is described in more detail below. Familiarisation.-Subjects entered the laboratory and were welcomed by the investigator. Subjects were given a brief description of the tasks to be performed and were asked to sign a statement of informed consent before taking part in the experiment. Subjects then completed the short self-report questionnaire before they practised regulating their level of exercise exertion using the W E . Subjects were instructed to use the 6-20 range of the W E to vary their work rate and become famhar with the scale. Subjects were also shown the F e e h g Scale and asked to indicate their present f e e h g to famiharise them with this scale. Trial 1.-Subjects were seated on an electronically braked cycle ergometer and requested to cycle at one of three different levels of W E (9, 13, or 17). They were requested to report their psychological affect in the last 20 sec. and 5 min. after performing a steady-state exercise bout of 5 min. duration. Oxygen uptake and heart rate were measured continuously using an on-line gas analysis and cardorespiratory system. Pedalling frequency was maintained at 55-60 revolutions per minute. Steady-state oxygen uptake and heart-rate values were used to predict the maximal oxygen uptake corresponding to the age-predicted maximal heart rate for each subject. The percentage maximal oxygen. uptake at each W E was subsequently calculated from this value. Subjects then had a 15-min. rest before commencing Trial 2. Trials 2 and 3.-The procedure for Trial 1 was repeated with subjects working at a different level of W E . The order of exercise was balanced across subjects to ensure that results were not confounded. RESu LTS Independent t tests were conducted on the biometric data of the sample. As indicated in Table 1, no significant differences were recorded for age, weight, or height, but significant differences were recorded for self-reported activity ( t= 9.18, p < .001) and predcted maximal oxygen uptake, V02max (t=4.30, p < .001). A mixed-model two-factor (group x WE) analysis of variance, with repeated measures on W E , was conducted on the calculated percentage rnaxima1 oxygen uptake measures (% V02max). The only significant effect was a main effect for W E (F,,5,=262.3 1, p < .001) with % VOzmax significantly increasing at each W E such that W E 9 = 3 1.3% V02max, WE 13 =57.9% V02max, and RPE 17 =79.7% V02max. This result does not support the first stated hypothesis that low-active subjects work at a lower proportion of the maximal oxygen uptake for a given W E . A mixed-model three-factor [group (high-active vs low-active) x time

1039

AFFECT. PEKCEIVED EXERTION

TABLE 1 MEANS AND STANDARD DEV~ATIONS OF SUBJEC~S' BIOMETRIC DATA Measure

H ~ g hActive M

Age, yr. 27.20 3.70 Self-report Activity, days 3.03 Estimated VOzrnax, Vmin. Body Mass, kg 60.60 156.8 Heieht. crn *Two groups are significantly different at p < ,001

-

Low Active

SD

M

6.99 1.28 0.41 6.56 43.3

25.60 0.33 2.29 61.23 165.8

SD 9.98 0.62* 0.53" 7.72 6.3

-

(-20 sec, vs + 5 min.) x W E (9, 13, and 17)] analysis of variance with repeated measures on time and W E indicated main effects ( p < .001) for each independent variable. High-active subjects reported more positive affect (M=2.12, SD = 1.6) than low-active subjects (M=0.88, SD = 1.7); subjects reported more positive affect at +5 min. (M=2.01, SD= 1.42) than at -20 sec. ( M = 0.91, SD = 1.87); and all subjects reported more positive affect as the

Rating of Perceived Exertion FIG.1. Interaction of Group [High-active women ( m ) , low-active women ( )I and WE with ratings on the Feeling Scale as the dependent variable. Values are mean _+ srandard error of the mean.

1040

G. PARFITT, ET AL.

W E was reduced ( W E 9, M=2.13; RPE 13, M = 1.65; and RPE 17, M = 0.69). Interactions were also recorded for group x RPE (F2,,=3.84, p < .003) and time x W E (F,,5,= 66.89, p < .001). Follow-up tests indicated that lowactive subjects felt less positive at RPE 17 than at W E 9 and less positive than the high-active subjects at W E s 9, 13, and 17 (see Fig. 1 ) . In addition, all subjects felt more positive 5 min. after exercising than in the last 20 sec. of exercising at W E s of 13 and 17 (see Fig. 2).

9

13

17

Rating of Perceived Exertion FIG. 2. Interaction of WE and time [last 20 sec. of exercise

(A

)I with r;lrlngs

error of the

on

( + ), 5 min. postexercise the Feeling Scale as the dependent variable. Values are mean f standard

mcJn

The main effect for RPE does not support the hypothesis or previous literature which indicates that the low-active subjects would work at a lower proportion of the maximal oxygen uptake for a given WE. The groups dld not differ in their predicted %V02max at each W E , and these predicted values indicated that the groups had been working at three different exercise work rates. Although previous work has shown that WE can be used in a self-regulation (production) protocol, these studies have not explicitly in-

AFFECT, PERCEIVED EXERTION

1041

cluded low-active women (see for example, Eston & Wdliams, 1988). Indeed, the studles on which the hypothesis was based had used trained and untrained subjects (Demello, et al., 1987; Boutcher, et al., 1989) and highactive and low-active men rather than women (Dishman, et d l . , 1994). The present results indcate that high- and low-active women can use RPE in a production protocol to elicit similar relative intensities of exercise at three distinct W E levels. This clearly has implications for health promotion where the target population group is generally a sedentary, low-active population. The ratings of affect reported in this study support previous findings by Parfitt and colleagues and have implications for the intensities of physical exertion recommended to low-active subjects and when subjects are encouraged to focus upon how they feel, if they are to feel good about exercising. Ln particular, it appears that the women in the low-active group felt relatively good about exercising at RPEs of 9 and 13 during which their scores were comparable to the high-active subjects. However, at RPE 17 the lowactive subjects felt significantly worse than at RPEs of 9 and 13, and were less positive than high-active subjects at RPE 17. This indicates that exercise leaders and health promoters could recommend that low-active subjects work at RPE 13 without being concerned that the experience will be interpreted very negatively and lead to exercise avoidance. An intensity of exercise corresponding to W E 13 should elicit a significant carlovascular irnprovement as the intensity of exercise elicited at W E 13 (58% in this study) is greater than the minimal threshold value (50% V02max) necessary for improving maximal oxygen uptake in normal healthy adults (American College of Sports Medicine, 1990). Further, the total amount of work done in a given period of time at RPE 13 would be greater than at W E 9. This is also considered to be a very important factor for the improvement of aerobic fitness (American College of Sports Medicine, 1990). An important consideration for prescribing the intensity of exercise through ratings of perceived exertion is that RPE at a given intensity of exercise appears to be mediated by the size of the muscle mass involved (Ekblom & Goldbarg, 1971). In addition, the W E is also medated by pedalling frequency (Robertson, Gillespie, McArthy, & Rose, 1979) although this latter factor was controlled in the present study. Thus, although W E appears to be higher at a given intensity of exercise for c y c h g compared to running (see Wdiams & Eston, 1996 for a discussion of this point), it is not known if psychological affect is similarly medlated by the mode of exercise. Further study is needed to determine whether the results from the relatively unfamiliar cycle ergometry exercise used in this study may be generaked to a recreational activity in which the individual is highly motivated to participate. The interaction of time x W E silggests that the negative effects of exer-

1042

G. PARFITT, ET AL.

cise dlstress felt during exercise may be precluded by having subjects focus upon postexercise affect. This is crucial at the h~gherintensities of exercise (RPEs 13 and 17) when irrespective of activity level, subjects reported more negative affect during exercise than after it had ceased. If it is necessary for subjects to work at WE 17, then it is advisable that they do not focus upon how they feel during the exercise bout. Subjects feel significantly more positive 5 min. after the exercise session has ended. REFERENCES AMERICAN COLLEGE O F SPORTS MEDICINE. (1990) The recommended quantity and quality of exercise for developing and maintaining cardiovascular and muscular fitness in healthy adults. Medicine and Science in Sports and Exercise, 22, 265-274. ~ T R A N D ,P. 0. (1988) Work tesfs zuith the bicycle ergormeter. Varberg, Sweden: Monark Crescent AB. BOUTCHER, S. H., SEIP,R. L., HETZLER, R. K., PIERCE,E. F., SNW, D., & WELTMAN, A. (1989) The effects of specificity of trainin on rating of perceived exertion at the lactate threshold. European Jozrrnal of Applied ~fysrology,59, 365-369. BORG,G. A. V. (1970) Perceived exertion as an indicator of somatic stress. Scarrdinavian Jozrrnal of Rehabilitation Medicine, 2-3, 92-98. DEMELLO, J. 1.. CURETON, K. , BOINEAV, R. E., & S I N G HM. , M. (1987) Ratings of erceived exertion at the lactate tkreshold in trained and untrained men and women. &d:'cine and Science in Sports and Exerciie, 19, 354-362. DISHMAN. R. K.. FARQUHAR. R. P.. & CURETON,K. J. (1994) Res onses to preferred intensities of exertion in men differing in activity levels. Medicine a n f ~ c i e n c ein Sports and Exercise, 26., 783-790. -. -. . . DUNBAR, C. C., ROBERTSON: R. J., BAUN,R., BUNDIN,M. F., METZ,R. K., BURDETT, R., &GOSS, E L. (1992) The validity of regulating exercise intensity by ratings of perceived exertion. Medicine and Science in Sports and Exerase, 24, 94-99. EKFJLOM, B., &GOLDBARG, A. N. (1971) The influence of ph sical training and other factors in the subjective rating of perceived exertion, Acta physioJgica Scandinavica, 83, 399-406. ESTON,R. G., DAVIES,B., &WILLIAMS, J. G. (1987) Use of perceived effort ratin s to control exercise intensity in young, healthy adults. Ezrropean Journal of Applied ~ t f ~ s i o l56, o~~, 222-224. of ratin s of perceived effort for the regulaorts ~ e j i c i n e22, , 153-154. ut how one feels: the measurement of affect during exercise. Jonrtzal of Sport and Exercise Psychology, 11, 304-3 17. LEVENTHAL, H., & EVERHART, D. (1979) Emotion, pain, and h sical illness. In C. E. Izard (Ed.), E~notionsin personality and psychopathology. New &rE: Plenum. Pp. 263-298. MOSES,J.. STEPTOE,A,, MATHEWS, A., &EDWARDS, S. (1989) T h e effects of exercise training on mental well-being in the normal population: a controlled trial. Journal of Psychosomatic Research, 33, 47-61. PARFITT, G., &ESTON,R. G. (1995a) Changes in rating of perceived exertion and psychological affect in the early stages of exercise. Perceptual and Motor Skills, 80, 2:9-266. PARFITT, G., &ESTON,R. G. (1995b) HOWto help low-active women feel better about exercisi n g In R. Vanfraechem-Raway & Y. Vanden Auweele (Eds.), lXth E~rropeanCongress on Sport Psychology. Leuvin: Belgian Federation of Sport Psychology. P p . 876-882. PARFITT,G.. MARKLAND, D., &HOLMES,C. (1994) Responses to physical exertion in active and inactive males and females. Journal of Sport and Exercise Psychology, 16, 178-186. REJESKI,W. 1.. BEST, D..GRIFFITH.P., & KENNEY,E. (1987) Sex-role orientation and the responses of men to exercise stress. Research Qzrarterly, 58, 260-264. ROBERTSON, R. J., GILLESPIE, R. L.. MCARTHY,J., &ROSE,K. D. (1979) Differentiated erceptions of exertion: Part 1. Mode of integration and regional signals. Perceptrral a n 8 ~ o t o r Skills, 49, 683-689. \VILLIAMS, J. G., & ESTON,R. G. (1996) Exercise intensity re ulation. In R. G. Eston & T. Reilly (Eds.), Kinan~hropomet~ and exercise physiology ?aboratory manual: test^, pmcedzires, and data. London: E. & F.N. Spon. Pp. 221-235.

Accepted April 5, 1996.