The role of relative autonomy in post-exercise affect responding.

Lutz, Rafer ; Lochbaum, Marc ; Turnbow, Kimberly 等

Exercise psychology researchers have sought to determine causal mechanisms responsible for affect in response to acute bouts of aerobic exercise. Though numerous mechanisms have been investigated such as the thermogenic hypothesis or Solomon's theory of acquired motivation, no one mechanism or theory has gained consistent empirical support (Petruzzello, Landers, Hatfield, Kubitz, & Salazar, 1991). Interestingly, though psychological variables have always been prominent in other exercise psychology research domains (e.g., exercise adherence), these psychological variables have been overlooked in regard to affective responding to an acute bout of exercise. A potential reason for this oversight is that researchers may have believed that affective responses to a "physical" activity (exercise) should have a corresponding physiological origin (Tuson & Sinyor, 1993). However, Tuson and Sinyor (1993) and more recently Ekkekakis and Petruzzello (1999) have suggested that motivational and psychological variables bear p articular need for investigation as determinants of exercise-induced affective change. Tuson and Sinyor have stated that, "quite conceivably it is not exercise per se that leads to changes in affect, but rather psychological variables related to the experience of engaging in exercise" (p. 114). Extraversion, sensation seeking, and self-efficacy beliefs, for example, have been identified as potential moderators or mediators of the exercise-affect relationship (Ekkekakis & Petruzzello, 1999). Though not identified by the latter authors in this regard, an exerciser's regulatory profile is a leading candidate for study and is the focus of this investigation.

Self-determination theory (SDT; Deci & Ryan, 1985, 2000) differentiates regulatory processes through which an individual pursues personal goals and activities. Participation in physical exercise is one such example of a personal goal. Deci and Ryan (2000) identified six regulatory processes that fall along the self-determined continuum (nonself-determined to fully self-determined): amotivation or non-regulation to external, introjected, identified, integrated and intrinsic regulation. Individual behaviors, therefore, are said to differ in the degree to which they are autonomous or self-determined. For instance, some behaviors are engaged in with a sense of personal choice or autonomy while external rewards or pressures control others. The former constitute intrinsic forms of motivation, while the latter constitute extrinsic motivational forms.

Pertinent to the present series of investigations, a central feature of this theory is that goal pursuit and behaviors will lead to positive or negative outcomes (such as improved affect) based on the extent to which they are more or less self-determined. This is theorized to occur because intrinsically regulated behaviors more fully satisfy the basic needs of competence, autonomy, and relatedness, and this need fulfillment contributes to a range of beneficial effects (Deci & Ryan, 1995, 2000). Additionally, behaviors originally motivated by extrinsic sources can become more autonomous by means of "organismic integration" (Deci & Ryan, 1995, P. 34). While an individual may, at first, pursue an activity to satisfy external pressures or to gain rewards, his or her reasons for engaging in this activity may become integrated into his or her self-concept, making it a more self-determined activity. Self-determined behaviors, therefore, are theorized to lead to a variety of positive outcomes because of their ability to provide basic need fulfillment and contribute effectively to self-concept (Deci & Ryan, 1995).

Vallerand and Rousseau (2001) contend that motivation for sport and exercise has affective consequences as well as cognitive and behavioral consequences. Additionally, they proposed that such consequences will be increasingly positive as an individual experiences more intrinsic, self-determined forms of motivation. According to Vallerand's (1997) hierarchical model of intrinsic and extrinsic motivation, there are three levels of generality at which an individual may be motivated in either a self- or nonself-determined manner. The broadest level, the global level, refers to a trait-like orientation to adopt intrinsic, extrinsic, or amotivated forms of motivation. The second level is referred to as the contextual level. This level refers to an individual's typical motivational profile in a given context, such as exercise environments. The most specific level is the situational level that refers to an individual's self-determined motivational pattern at a specific point in time or with respect to a specific act ivity.

In support of Vallerand's (1997) and Deci and Ryan's (1985,2000) theorizing, self-determined motivation has been linked to psychological well-being and mental health (Ryan, Rigby, & King, 1993), persistence of motivated behaviors (Ryan, Frederick, Lepes, Rubio, & Sheldon, 1997), and well-being, positive affect, and vitality (Sheldon, Ryan, & Reis, 1996). More specifically, Sheldon and colleagues (1996) found that trait perceptions of autonomy were positively and significantly (p < .05) related to total well-being (r = .36), positive affect (r .28), and vitality (r = .34). Most recently, Sheldon, Elliot, Kim, and Kasser (2001) found that positive affect during participants' most satisfying events was predicted by ratings that this event met needs for autonomy (r range = .31 to .51), competence (r range = .39 to .59) and relatedness (r range = .21 to .33), and that these needs may be among a fundamental or core set necessary for human satisfaction. Considering these findings, it is logical to speculate that th e reasons! goals an individual has for engaging in exercise may influence his or her affective response to its completion.

Despite the theoretical appeal, only a handful of studies have been conducted to examine the relation between positive affective outcomes in sport and exercise and intrinsic motivation (Blanchard & Vallerand, 1996, 1998, as cited in Vallerand & Rosseau, 2001; Frederick, Morrison, & Manning, 1996; McAuley & Tammen, 1989). Blanchard and Vallerand (1996), as detailed by Vallerand and Rosseau (2001), found that positive affective responses immediately after a basketball game were positively related to intrinsic and identified forms of motivation. In another unpublished investigation, the authors reported that intrinsically motivated individuals who engaged in an exercise-based weight loss program reported greater levels of experienced positive emotion(Blanchard & Vallerand, 1998, as cited by Vallerand & Rosseau, 2001). Additionally, McAuley and Tammen (1989) found that successful basketball players reported greater levels of enjoyment, effort, and competence, indicating greater levels of intrinsic motivation. Fr ederick and colleagues (1996) found that exercisers who reported greater levels of competence, enjoyment, and social satisfaction (indicative of intrinsic motivation) reported greater enjoyment attributable to exercise. Though limited in number, these studies conducted in sport settings support the idea that affective response to exercise may be, in part, moderated by an individual's self-determined motivational profile. Yet, to further corroborate this link, investigations are needed to test these relationships in exercise contexts specifically.

Hence, the purpose of the present series of investigations was to test the predictions of SDT and the hierarchical model of intrinsic and extrinsic motivation specific to affective consequences attributable to exercise participation. The present studies also uniquely add to the extant literature in that they utilized a self-determined motivation measure developed specifically for assessment in exercise settings, and used affect measures commonly used in the research examining exercise-affect relationships. None of the aforementioned investigations have employed these measures.

EXPERIMENT 1

The primary goal of Experiment 1 was to determine if perceived typical affective response to exercise could be predicted by participants' self-determined motivational orientations. To achieve this end, we asked participants to fill out the Exercise Motivation Scale (EMS) a context-specific measure of self-determined motivation in exercise (Li, 1999). Subsequently, they were asked to engage in a brief imagery exercise designed to elicit their typical, contextual reaction to exercise participation. In this manner, we sought to determine if the contextual measure of motivation could predict contextual consequences of exercise as Vallerand's (1997) model would predict. The examination of the contextual measure of motivation allowed for the testing of several hypotheses.

Hypothesis one posited that mean values for intrinsic motivation types, integrated and identified regulation will be greater than those for less identified forms of motivation. This pattern of motivation has been found using this scale in college-student exercisers (Li, 1999) and using a related scale to examine sport participants (Pelletier, Fortier, Vallerand, Tuson, Briere, & Blais, 1995). Conducting a t-test comparing the mean of intrinsic motivation types, integrated regulation and identified regulation with the mean of introjected regulation, external regulation and a motivation tested this hypothesis. Hypothesis two predicted that positive correlations would be found between the three intrinsic forms of motivation, integrated regulation, identified regulation and the dependent variable of interest, positive affect. Conversely, negative relationships were predicted between the other motivational forms (introjected and external regulation, and amotivation) and positive affect. Last, we predicted that the degree to which an individual reported self-determined motives for exercise participation would positively relate to affective outcomes. To quantify self-determination, Deci and Ryan's (2000) Relative Autonomy Index (RAI, called the Self-Determination Index as described in Vallerand, 1997) was computed, and correlating the RAI with positive affect tested this relationship.

Method

Participants

Participants were 141 volunteer, college-aged university students (71 male, 70 female). All participants were recruited via personal communication from introductory psychology courses at a large southwestern university. Participants received credit towards completion of research participation required for successful completion of the introductory psychology course. Descriptive data are presented in Table 1.

Measures

EMS. The EMS includes 31 potential reasons or lack of reasons for engaging in exercise participation. Participants are asked to indicate their agreement with each statements as a reason why they exercise on a Likert-type scale ranging from 1 "strongly disagree" to 6 "strongly agree." This scale was developed specifically to assess self-determined motivational orientations in exercise, as opposed to sport, settings. It has demonstrated adequate levels of internal consistency [alpha] = .71 to .90), test-retest reliability (r =.78 to .88), and includes all eight dimensions along the self-determination continuum. Other SDT-based measures developed for use in sport or exercise settings have not identified all eight dimensions acceptably (Li, 1999).

Positive Affect Negative Affect Schedule (PANAS). The PANAS developed by Watson, Clark, and Tellegen (1988) includes 20 adjective words, 10 describing negative mood states and 10 describing positive mood states. The response set ranges from 1 "very slightly or not at all" to 5 "extremely." The internal consistency alphas range from .86 to .90 for positive affect (PA) and from .84 to .87 for negative affect (NA) (Watson et al., 1988). Participants were asked complete the adjective checklist in response to their feelings immediately after their imagined acute bout of exercise. In the present investigation, only the PA subscale was scored.

Participants were asked to complete a survey requiring approximately 45 minutes of their time. Within this exercise-focused survey, participants completed the EMS and an imagery session concerning exercise and affect. Participants were instructed to imagine completing a 30-minute exercise session. The mode and intensity of exercise were self-chosen. After choosing their exercise, participants were asked to take a few minutes to imagine completing this session. Next, participants were instructed to immediately complete the PANAS with regards to the imaged exercise session.

Results

Prior to analyzing our main hypotheses, we first determined whether any gender differences existed for either the EMS or affective reporting. A series of one-way ANOVAs revealed that no significant (p < .05) differences existed between males and females.

To examine our first hypothesis, that participants (who have self-selected to participate in a fitness based class) would have intrinsic motivation and integrated and identified regulation as their primary motivational orientation, the means were examined by conducting a dependent t-test comparing the mean of the self-determined subscales (M [+ or -] SD 4.27 + .80) with the mean of the non-self-determined subscales (M [+ or -] SD 2.44 + .79). The results, t(1, 141) = 20.79, p < .001, verified this hypothesis. This, in addition to visual inspection of the pattern of means, confirmed the hypothesized relationship (see Table 1). Simple correlations (see Table 2) were computed to examine our second and third hypotheses concerning the relationship among EMS subscales, RAI, and exercise-induced affect. As seen in Table 3, participants reported a fairly high level of positive affect in response to the imagery session concerning their self-chosen mode and intensity of 30 minutes of exercise. The simple correlations with positive affect were significant (p < .05) and in the hypothesized direction for identified and integrated regulation, each type of intrinsic motivation, and the RAT. Though negative in direction, the correlations between amotivation, external regulation and positive affect were not statistically significant. Thus, the modest to moderate magnitude correlations generally supported our hypotheses with regards to the proposed relationships along the self-determination continuum with positive affect reporting in response to an acute bout of exercise.

EXPERIMENT 2

Experiment 2 was conducted to replicate the findings of Experiment 1 within a setting that would provide greater ecological validity; and, additionally, to address several limitations inherent in the design of the first investigation. In particular, we sought to measure participants' affective response to exercise using a paradigm more typical to this line of investigation (i.e., directly after an actual exercise bout); implement the use of additional measures of affect; control for effects of pre-exercise affect; and reduce testing effects by separating the presentation of the EMS and affective measures in time. Therefore, Experiment 2 explored whether contextual motivational orientations could predict situational affective response to exercise.

Participants completed the EMS two days before they returned to their scheduled class meeting time to engage in a bout of aerobic exercise. Affect was measured with two scales currently receiving popular use at three time points: immediately before exercise, immediately after exercise, and 15 minutes after completion of exercise. In this manner, entering pre-exercise affect into a regression as a predictor simultaneously with RAT to predict post-exercise affect would control for the effects of participants' trait or state dependent affect (i.e., affect in anticipation of an exercise session, affect as a result of stimulus prior to class, or trait-orientations to experience positive affect). Finally, this set-up allowed for an additional test of Vallerand's (1997) hierarchical model by determining whether motivation at the contextual level influences situational outcomes.

In Experiment 2, the predictions made by Hypotheses 1-3 in the first investigation were retained and retested. In order to determine whether pre-exercise affect influenced post-exercise affective reports, and to control for potential influences of global or contextual motivational orientations that may influence contextual affect (participants knew they would be exercising) or global affect (trait autonomy has been shown to relate to general well-being, Sheldon et al., 1996), a fourth hypothesis was advanced. This fourth hypothesis predicted that contextual motivation would contribute unique variance to the prediction of post-exercise affect above and beyond pre-exercise affect.

Method

Participants

Participants were 99 volunteer, college-aged university students. All participants were female and recruited via personal communications from aerobic and/or fitness courses at a large southwestern university. Descriptive data are presented in Table 1.

Measures

Activation Deactivation Adjective Checklist (AD ACL). In addition to utilizing the PANAS as in Experiment 1, the AD ACL (Thayer, 1989, Appendix A) was used to assess affect. The AD ACL is a 20-item self-report inventory that assesses the following two-arousal dimensions: energetic arousal (EA) and tense arousal (TA). EA and TA are compatible to dimensions of positive activation (positive affect) and negative activation (negative affect) in Watson, Wiese, Vaidya, and Tellegen's (1999) model, respectively. In the present investigation, only EA was scored. The AD ACL's reliability and construct validity are well established (Thayer, 1986).

Ratings of Perceived Exertion (RPE). RPE was completed based on Borg's (1985) 10-point graded scale prior to, during, and post exercise. This scale is a self-report, psychophysical measure of perceived effort. The scale ranges from 0 "nothing at all" to 10 "very, very strong" to a single point of maximal. RPE as measured in a variety of graded scales has been shown to be a valid and reliable indicator of relative fatigue for over 30 years (Pollock & Wilmore, 1990).

Procedure

After receiving permission from the class instructor to perform research in aerobics classes, participants were approached within the class period and provided an overview of the investigation. Then they were given the approved informed consent as approved by the second author's University Human Subject Board to read and sign. The investigation required the participants to complete a survey that contained the EMS and demographic information. After completion of the survey and during a different class period, participants completed a 30-minute step aerobic class. Participants completed ratings of perceived exertion and affect (PANAS and AD ACL) prior to, during cool down, and 15 minutes after completion of the aerobic session.

Data Analyses

First, we conducted repeated-measures multivariate analyses of variance on the affect scores to verify, as would be predicted, that participants reported increased positive affect from baseline values. To examine our major hypotheses, simple correlations among the EMS facets, RAI, and positive affect were examined. Next, regression analyses were conducted to determine the predictive power of RAI over and above that of pre-exercise affect.

Due to logistical constraints (i.e., time constraints and the wishes of the instructor for the given class periods), 37 of the 99 participants failed to complete the PANAS during any of the three time points and the AD ACL at the delayed post-test. For the analyses examining pre-and post- AD ACL, therefore, data from all study participants (N = 99) were entered into the analyses. The analyses examining the PANAS and delayed AD ACL delayed post-test utilized the data from the 62 participants who had complete data sets.

Results

To verify the similarity of intensity of the aerobic sessions, RPE was collected mid-way through the aerobic session and also at the two time points during recovery. Participants reported statistically similar (p> .05) RPE values (M [+ or -] SD, During: 5.43 2.06, 5.00 [+ or -] 1.94; Post: 5.88 2.33, 5.47 2.25; Delayed post: 4.03 1.45) that corresponded to strong (moderate intensity) on the Borg scale in response to the regardless their grouping, n = 62 or n 99, respectively.

To determine whether self-reported affect followed expected temporal patterns, a series of repeated measure MANOVAs were conducted. In each analysis, the Huynh-Feldt epsilon was examined and multivariate statistics were used when [epsilon] < .75. For the 62 participants who completed the PANAS and AD ACL at all measurement time points (pre, post, and delayed post), the main effect for Time was significant, F(2, 122) 4.17, [epsilon] = .79, p <.05. Univariate followup tests were conducted to determine which dependent variables were contributing to the multivariate Time effect, and there were significant Time main effects for both PA (PANAS), F(2,60) = 19.87, p <.05, and for EA (AD ACL), F(2,60) = 25.32, p <.001. Finally, the data for the 37 participants who failed to complete the PANAS and the delayed AD ACL were examined in addition to the 62 who had complete data. These 99 participants had completed the AD ACL at pre and posttest, and the main effect for Time was significant, Wilks' [lambda] = .83, F( 1,98) = 20.34, p <.001. In all analyses, inspection of the means suggested that the main effects for Time were due to increased and generally sustained positive affect reporting (compared to pre-exercise positive affect) in response to the bout of aerobic exercise (see Table 3).

As in Experiment 1, to examine our first hypothesis that participants who have self-selected to participate in a fitness based class would have intrinsic motivation and integrated and identified regulation as their primary motivational orientation, a dependent t-test was conducted on the means corresponding to self-determined (M [+ or -] SD = 4.24 [+ or -] .84) and non-self-determined motivation (M [+ or -] SD = 2.38 .65) and verified hypothesis one, t(1, 98) = 16.28, p < .001. In addition, visual inspection of the pattern of means confirmed the hypothesized relationship (see Table 1). Simple correlations (see Table 2) were examined to test our second and third hypotheses concerning the relationship among EMS subscales, the RAI, and exercise-induced positive affect. The simple correlations were significant (p < .05) and in the hypothesized direction for all relationships. Thus, the modest to moderate magnitude correlations with positive affect strongly supported our hypotheses with regards to relationships al ong the continuum of amotivation to intrinsic motivation. Last, to examine the predictive ability of the RAI over and above that of pre-exercise affect, hierarchical regression analyses were performed (see Table 4). The RAI significantly (p <.05) predicted unique variance (13 and 6%) in affect immediately after exercise and also significantly predicted affect (4%) 15 minutes after exercise completion for the PANAS.

Discussion

The results of the present series of investigations clearly suggest that self-determined motivational orientations have an impact on positive feelings reported by participants post-exercise. Additionally, and in support of Vallerand's (1997) hierarchical model of intrinsic and extrinsic motivation, this research indicates that contextual motivation appears to influence situational consequences. Finally, this research supports the exercise adherence literature, which proposes that enjoyment and intrinsic motivation are important elements in the adoption and maintenance of habitual exercise programs.

The present series of investigations uniquely contribute to the exercise psychology literature for several reasons. First, few studies have examined motivational variables and the influence they may have on the affective response to exercise (Blanchard & Vallerand, 1996, 1998, as cited by Vallerand & Rosseau, 2001; Frederick et al., 1996; McAuley & Tammen, 1989; Tate, Petruzzello, & Lox, 1995; Treasure & Newbery, 1998). Next, even fewer studies have examined self-determined motivational types in this respect (Blanchard & Vallerand, 1996, 1998 as cited by Vallerand & Rosseau, 2001; Frederick et al., 1996). Finally, seldom have these studies examined affective change after exercise bouts using the typical paradigm (i.e., affective pretest, acute exercise bout, and posttests with various delay periods). Though Treasure and Newbery (1998) and Tate et al. (1995) have examined self-efficacy states as predictors of affective response using similar procedures, no known study has examined motivational variables derive d from SDT in this manner. Two studies examining SDT in this area have examined affective or enjoyment responses after sport participation (Blanchard & Vallerand, 1996, as cited by Vallerand & Rosseau, 2001; McAuley & Tammen, 1989), and the one known study examining affective response to exercise failed to control for pre-exercise affect and did not assess situational affect, but rather contextual affect (Frederick et al., 1996). In this manner, Experiment 1 corroborates the findings of Frederick et al., and Experiment 2 extends their work by including situational outcomes. Though researchers have focused on physiological explanations for affective changes attributed to exercise, clearly a portion of the variance in these situations can be explained by psychological/motivational variables. It may be of interest to determine how physiological variables might ultimately influence psychological responses and vice-versa.

Of the studies that closest approximate the present set of investigations, Tate and colleagues (1995) found that self-efficacy states before exercise predicted EA at 30 minutes post-moderate intensity exercise (55% of VO2max). This corroborates the findings of the present investigation indicating that positive affective outcomes ascribed to moderate intensity exercise in the present investigation may be, at least partially, explained by motivational variables. It is of interest, in the future, to determine which constructs (self-efficacy, self-determination or both) have the primary influence upon positive affect in exercise situations. Also, similar to the present investigation, Frederick et al. (1996) found that self-reported high adherers to physical activity reported greater competence, enjoyment, and social satisfaction due to exercise indicating a more self-determined motivational profile in comparison to low adherers. In addition, ratings of competence, enjoyment, and social motives correlated positive ly with post-workout ratings of enjoyment. Frederick and colleagues' findings support the present results and suggest a link between motivation, exercise-related affect, and exercise participation. Future research should examine the meaningfulness and intricacies of all of these relationships concurrently.

The present research extends the research conducted by Frederick and colleagues (1996) by using a measure of intrinsic and extrinsic motivation developed specifically for exercise contexts, examining affect directly after an actual exercise bout, using affect measures commonly used to examine exercise-induced affective changes, and controlling for pre-exercise affect (Frederick and colleagues only examined contextual enjoyment due to exercise). In the present research, situational affect (Experiment 2) was measured in addition to a contextual affect (Experiment 1). In both cases, contextual motivational orientations had an influence on affective reports, thus corroborating Vallerand's (1997) theorizing that contextual motivation can have both contextual and specific consequences. This is noteworthy because this is the first known research to examine consequences following exercise within Vallerand's framework using the traditional paradigm of the study of affective responses to acute exercise.

It is of interest to note the relation between contextual autonomy (RAI) and pre-exercise affect in Experiment 2 (EA: r .24; PA: r .47). It follows the theorizing by Vallerand (1997) that contextual motivation for exercise influenced situational affect (i.e., participants who are autonomously motivated exhibited more pre-exercise positive feelings knowing they were going to be exercising). It may also be, however, that global (trait) self-determination had an influence on situational affect. Higher trait autonomy levels are positively correlated with daily wellbeing (Sheldon et al., 1996). Sheldon and colleagues found that trait (global) and state (situational) feelings of competence or autonomy predicted positive feelings. The relation between contextual motivation and situational affect at the pre-test may, in part, be due to similar, positive effects of global orientations on both variables. This would be predicted by Vallerand's model as well. It remains for future studies, however, to determine any diffe rential effects of global, contextual, and situational motivation on "average" well-being and situational positive affect after exercise. The results of the present investigation do confirm that contextual motivation appears to have an influence on both pre- and post-exercise affect. This finding has strong implications for the manner in which affective changes due to exercise is 'currently studied. Depending on the nature of the relationship, exercise participation may influence affective response before exercise, in addition to post exercise. The correlation between contextual autonomy and pre-exercise affect suggests that it may be important to study affective response to exercise within the context of an individual's average daily affect.

Similar to Frederick and colleagues' (1996) study, other researchers have found that exercise participation and adherence appear to be related to autonomous motivation (Oman & McAuley, 1993; Ryan et al., 1997). Oman and McAuley (1993) noted that different goals cause an individual to begin an exercise program (usually health benefits) in comparison to the goals with the goals that lead to continued involvement in such a program (such as social reasons). Oman and McAuley described how situations may influence the level of intrinsic motivation.

Exercisers that pursue fitness for controlled reasons should exhibit less intrinsic motivation than those that pursue fitness for less controlling reasons. Thompson and Wankel (1980) is an example of a study where this relationship is supported. They found that exercisers who perceived more choice in the physical activities they pursued were more likely to exhibit feelings of self-determination, intrinsic motivation and better program attendance. Oman and McAuley (1993) further examined the relation between intrinsic motivation and exercise behavior by administering the Intrinsic Motivation Inventory (IMI) to community aerobics exercisers. It was found that participants with higher intrinsic motivation scores attended more exercise classes. Also, participants' confidence in their intentions to continue to exercise after the end of the exercise program was related to intrinsic motivation scores measured before program initiation. These findings, in addition to the present findings, suggest that motivation infl uences affect and participation. Whether affect mediates that relationship between self-determined motivation and exercise participation remains to be studied.

Overall, the present research confirms Tuson and Sinyor's (1993) suggestion that exercise psychologists should broaden their scope to include psychological/motivational variables in the investigation of the exercise-affect relationship. Self-determined motivational types predicted both contextual and situational positive affect in exercise settings. The present studies are limited by the use of small, college-age samples that appear to exhibit a high level of autonomous regulation for exercise. Nonetheless, these studies confirm the predictions of SDT (Deci & Ryan, 1985, 1995, 2000) and the hierarchical model of intrinsic and extrinsic motivation (Vallerand, 1997), and extend our understanding of the exercise-affect relationship.

Table 1

Variable Means and Standard Deviations for Participants in Both Studies

 Experiment 1 Experiment 2
 (n=141) (n=99)
Variable Mean SD Mean SD

Descriptive
 Age 20.14 2.94 20.46 2.32
EMS subscales
 Amotivation 1.64 0.88 1.69 0.87
Extrinsic Motivation
 External Regulation 2.18 1.10 2.08 0.86
 Introjected Regulation 3.52 1.27 3.38 1.06
 Identified Regulation 4.84 0.91 4.92 0.87
 Integrated Regulation 4.15 0.90 4.18 0.96
Intrinsic Motivation
 To learn 3.41 1.24 3.43 1.13
 To accomplish things 4.40 0.98 4.21 0.96
 To experience sensations 4.58 0.97 4.46 0.99
 Relative Autonomy Index 12.75 6.54 12.78 7.48

Note. EMS = Exercise Motivation Scale.

Table 2

Correlations Among Post Exercise Positive Affect and EMS Subscales and
Total RAI Score

 Experiment 1 Experiment 2

 Post Imagine Pre
 PA PA EA
n 141 62 99

EMS subscales
Amotivation -.09 -.38 ** -.24 *

Extrinsic Motivation
 External Regulation -.08 -.31 ** -.15
 Introjected Regulation .19 * -.11 .05
 Identified Regulation .43 *** .32 * .14
 Integrated Regulation .39 *** .35 ** .13

Intrinsic Motivation
 To learn .21 * .25 * .11
 To accomplish things .42 *** .37 ** .21 *
 To experience .42 *** .39 ** .21 *
 sensations
 Relative Autonomy .36 *** .47 *** .24 *
Index

 Experiment 2

 Post Delayed Post
 PA EA PA EA
n 62 99 62 62

EMS subscales
Amotivation -.44 *** -.33 ** -.38 ** -.32 *

Extrinsic Motivation
 External Regulation -.40 ** -.22 * -.38 ** -.17
 Introjected Regulation -.15 .03 -.12 .05
 Identified Regulation .32 * .39 *** .34 ** .29 *
 Integrated Regulation .38 ** .35 *** .38 ** .20

Intrinsic Motivation
 To learn .26 * .31 ** .23 .21
 To accomplish things .41 ** .42 *** .42 ** .33 **
 To experience .31 * .41 *** .30 * .24
 sensations
 Relative Autonomy .52 *** .45 *** .49 *** .34 **
Index

Note. EMS = Exercise Motivation Scale; RAI = Relative Autonomy Index; PA
Positive Affect (PANAS); and EA Energetic Arousal (AD ACL). *** p <.001;
** p<.01; * p <.05 Procedure

Table 3

Variable Means (Standard Deviations, Effect Sizes, and Sample Size for
Self-Reported Affect for Both Studies)

 Experiment 1 Experiment 2

Variable Post Imagine Pre Post

PANAS
 Positive Affect 38.23 (7.34) 25.89 (9.42) 28.29 (9.68)
 ES -- -- .25
 n 141 62 62
AD ACL
 Energetic Arousal 23.29 (7.29) 29.61 (7.24)
 ES -- -- .87
 n 62 62
 Energetic Arousal 23.36 (6.44) 26.72 (7.24)
 ES -- .52
 n 99 99
Ratings of Perceived 5.04 (1.98) 5.54 (2.26)
 Exertion

 Experiment 2

Variable Delayed Post

PANAS
 Positive Affect 27.90 (10.03)
 ES .21
 n 62
AD ACL
 Energetic Arousal 28.50 (7.53)
 ES .71
 n 62
 Energetic Arousal
 ES
 n
Ratings of Perceived 4.12 (1.64)
 Exertion

Note. ES = effect size.

Table 4

Regression Analyses

 Model Results

 [R.sup.2] [R.sup.2.sub.adjusded]

Criterion Variable:
Immediate Post
AD ACL (a)
 Pretest .17 .17
 RAI .31 .29
PANAS (c)
 Pretest .41 .40
 RAI .47 .45

Criterion Variable:
Delayed Post
AD ACL (e)
 Pretest .33 .32
 RAI .37 .35
PANAS (g)
 Pretest .47 .46
 RAI .50 .49

 Model Results Coefficient Results

 [R.sup.2.sub.change] Beta t

Criterion Variable:
Immediate Post
AD ACL (a)
 Pretest .17 .33 3.76
 RAI .13b .38 4.28
PANAS (c)
 Pretest .41 .51 4.74
 RAI .06 (d) .28 2.57

Criterion Variable:
Delayed Post
AD ACL (e)
 Pretest .33 .52 4.89
 RAI .04 (f) .20 1.86
PANAS (g)
 Pretest .47 .58 5.59
 RAI .04 (h) .21 2.05

 Coefficien
 t Results

 p

Criterion Variable:
Immediate Post
AD ACL (a)
 Pretest .000
 RAI .000
PANAS (c)
 Pretest .000
 RAI .012

Criterion Variable:
Delayed Post
AD ACL (e)
 Pretest .000
 RAI 0.68
PANAS (g)
 Pretest .000
 RAI .045

Note. (a)Final model, F(2,96) = 21.22, p<.001;

(b)[F.sub.change] (1,96) = 18.35, p <.001;

(c)Final model, F(2,59) = 26.19, p <.001;

(d)[F.sub.change] (1,59) = 6.65, p <.05;

(e)Finl model, F(2,59) = 17.45, p <.001;

(f)[F.sub.change] (1,59) = 3.46, p <.05;

(g)Final model, F(2,59) = 29.72, p <.001;

(h)[F.sub.change] (1,59) = 4.19, p <.05.

References

Blanchard, C., & Vallerand, R. J. (1996). On the relations between situational motivation and situational consequences in basketball. Unpublished manuscript, University of Quebec, Montreal, Canada.

Blanchard, C., & Vallerand, R. 1. (1998). On the relations between situational motivation and situational consequences toward exercise. Unpublished manuscript, University of Quebec, Montreal, Canada.

Borg, G. (1985). An introduction to Borg's RPE scale. Ithaca, NY: Mouvement Publications.

Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self-determination in human behavior. New York: Plenum Press.

Deci, E. L., & Ryan, R. M. (1995). Human autonomy: The basis for true self-esteem. In M. H. Kemis (Ed.), Efficacy, agency, and self-esteem. (pp. 31-49). New York: Plenum Press.

Deci, E. L., & Ryan, R. M. (2000). The "what" and "why" of goal pursuits: Human needs and the self-determination of behavior. Psychological Inquiry 11, 227-268.

Ekkekakis, P., & Petruzzello, S. J. (1999). Acute aerobic exercise and affect: Current status, problems and prospects regarding dose-response. Sports Medicine, 28(5), 337-374.

Frederick, C.M., Morrison, C., & Manning, T. (1996). Motivation to participate, exercise affect, and outcome behaviors toward physical activity. Perceptual and Motor Skills, 82, 691-701.

Li, F. (1999). The exercise motivation scale: Its multifaceted structure and construct validity. Journal of Applied Sport Psychology, 11, 97-115.

McAuley E., & Tammen, V. V. (1989). The effects of subjective and objective competitive outcomes on intrinsic motivation. Journal of Sport and Exercise Psychology 11, 84-93.

Oman, R., & McAuley, E. (1993). Intrinsic motivation and exercise behavior. Journal of Health Education, 24, 232-238. Pelletier, L. G., Fortier, M. S., Vallerand, R. J., Tuson, K. M., Briere, N. M., & Blais, M. R. (1995). Toward anew measure of intrinsic motivation, extrinsic motivation, and amotivation in sports: The Sport Motivation Scale (SMS). Journal of Sport and Exercise Psychology 17, 35-53.

Pelletier, L.G., Fortier, M.S., Vallerand, R.J., Tuscon, K.M., Briere, N.M. & Blais, M.R. (1995) Toward a measure of intrinsic motivation, extrinsic motivation and amotivation in sport: The Sports Motivation Scale (SMS). Journal of Sport and Exercise Psychology 17,35-53.

Petruzzello, S.J., Landers, D.M., Hatfield, B.D., Kubitz, K.A., & Salazar, W. (1991). A meta-analysis on the anxiety reducing effects of acute and chronic exercise. Sports Medicine, 11, 143-182.

Pollock, M. L., & Wilmore, J. H. (1990). Exercise in health and disease. Philadelphia: W. B. Saunders.

Ryan, R. M., Fredrick, C. M., Lepes, D., Rubio, N., & Sheldon, K. M. (1997). Intrinsic motivation and exercise adherence. International Journal of Sport Psychology 28, 335-354.

Ryan, R. M., Rigby, S., & King, K. (1993). Two types of religious internalization and their relations to religious orientations and mental health. Journal of Personality and Social Psychology 65,586-596.

Sheldon, K. M., Elliot, A. J., Kim, Y, & Kasser, T. (2001). What is satisfying about satisfying events? Testing 10 candidate psychological needs. Journal of Personality and Social Psychology 80,325-339.

Sheldon, K. M., Ryan, R. M., & Reis, H. T. (1996). What makes for a good day? Competence and autonomy in the day and in the person. Personality and Social Psychology Bulletin, 22,1270-1279.

Tate, A. K., Petruzzello, S. J., & Lox, C. L. (1995). Examination of the relationship between self-efficacy and affect at varying levels of aerobic exercise intensity. Journal of Applied Social Psychology, 25, 1922-1936.

Thayer, R. E. (1986). Activation-deactivation adjective checklist: Current overview and structural analysis. Psychological Reports, 58, 607-614.

Thayer, R. E. (1989). The biopsychology of mood and arousal. New York: Oxford University Press.

Thompson, C. E., & Wankel, L. M. (1980). The effects of perceived activity choice upon frequency of exercise behavior. Journal of Applied Social Psychology, 10, 436-443.

Treasure, D.C., & Newbery, D. M. (1998). Relationship between self-efficacy, exercise intensity, and feeling states in a sedentary population during and following an acute bout of exercise. Journal of Sport and Exercise Psychology, 20, 1-11.

Tuson, K. M., & Sinyor, D. (1993). On the affective benefits of acute aerobic exercise: Taking stock after twenty years of research. In P. Seraganian (Ed.), Exercise psychology: The influence of physical exercise on psychological processes (pp. 122-145). New York: Wiley.

Vallerand, R. J. (1997). Toward a hierarchical model of intrinsic and extrinsic motivation. In M. P. Zanna (Ed.), Advances in experimental social psychology (pp. 271-360). New York: Academic Press.

Vallerand, R. J., & Rosseau, F. L. (2001). Intrinsic and extrinsic motivation in sport and exercise: A review using the Hierarchical Model of Intrinsic and Extrinsic Motivation. In R. N. Singer, H. A. Hausenbias, & C. M. Janelle (Eds.), Handbook of sport psychology (pp. 389-414). New York: Wiley.

Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54, 1063-1070.

Watson, D., Wiese, D., Vaidya, I., & Tellegen, A. (1999). The two general activation systems of affect: Structural findings, evolutionary considerations, and psychobiological evidence. Journal of Personality and Social Psychology, 76, 820-838.

Address Correspondence To: Marc Lochbaum, Ph.D., Department of Heath, Exercise, and Sport Sciences, Texas Tech University, Box 43011, Lubbock, TX 79409-3011. Telephone: (806) 742-3371. Fax: (806) 742-1688. E-mail: [email protected]