Introduction

Aerobic exercise prescription has been based on recommendations of international associations and independent authors [1,2,3,4,5]. For example, in its last guidelines, the American College of Sports Medicine (ACSM) [3] recommended that healthy individuals should perform an exercise session at an intensity between 30% and 89% of heart rate reserve (HRR) or reserve oxygen consumption (VO2R) and duration between 20 and 60 min. These recommendations suggest exercise sessions with a continuous and fixed intensity adjusted using different variables such as heart rate (HR) and oxygen consumption (VO2).

Despite the range available in the guidelines, it has already been demonstrated that an exercise performed at fixed intensity results in greater difficulty in maintaining the steady-state than an exercise performed at the self-selected intensity, even if performed with a similar intensity [6]. In this regard, Johnson and Phipps [7] reported that an individual could under or overestimate the pace when self-select the exercise session. Based on this argument, it is possible to hypothesize that individuals with lower cardiorespiratory fitness, and consequently, lower experience in aerobic exercise, are more susceptible to making mistakes in self-selection than individuals with higher cardiorespiratory fitness and experience. Corroborating the fact, Rose and Parfitt [8] showed that active women performed self-selected exercise at a higher percentage of the HR than sedentary women. However, the participants did not distinguish according to the fitness level, which could have contributed to these results.

Despite the aforementioned aerobic exercise recommendations, studies have suggested that regular exercise is still below the necessary [9,10,11,12]. Therefore, other exercise prescription strategies with a focus on exercise adherence should be tested. A relationship between adherence to aerobic exercise and positive affective responses has already been demonstrated [13, 14]. The affective response may be interpreted as a combination of valence (pleasure or displeasure), and arousal (low or high activation) frequently experienced [15]. In this context, Ekkekakis [16] suggested that aerobic exercise sessions performed at self-selected intensity (according to the practitioner's preferences) could result in positive affective responses during exercise and, therefore, may be a strategy for increase exercise adherence.

In this sense, the lactate and ventilatory threshold seem to play a central role in regulating self-selected exercise [17], considering that depending on the aerobic exercise intensity, below or above the lactate/ventilatory threshold, the maintenance of the steady-state may be compromised. In this respect, Cabanac and LeBlanc [18] demonstrated that individuals self-selected the exercise intensity to maintain a steady-state condition, which may maximize pleasure or minimize displeasure.

On the one hand, exercise sessions performed at fixed intensity (such as the formal recommendations of aerobic training prescription) may result in negative affective responses compared to self-selected exercises [6]. On the other hand, it has previously been argued that an individual can under or overestimate the pace when self-select the exercise session [7]. However, Johnson, Phipps [7] showed that individuals might perform self-selection within a safe intensity. However, no study differentiated the participants according to the cardiorespiratory fitness, which could have contributed to these results, considering the differences in individuals' exercise experiences with lower and higher conditioning.

Therefore, the objective of the present study was to compare the intensity and duration selected by participants with higher and lower cardiorespiratory fitness to that recommended by the ACSM [3]. Also, a comparison of affective responses between higher and lower conditioned individuals was conducted to verify the effect of self-selection and physical fitness on affective responses. The hypothesis of the present study was that participants with lower cardiorespiratory fitness would perform intensity and duration different from that recommended by the ACSM [3].

Methods

Procedure

Participants performed two visits to the laboratory with an interval between two and seven days. In the first visit, participants were asked to complete the stratification risk questionnaire [3]. Anthropometric measurements (body mass, height, and skinfold thickness) and resting HR, and blood pressure were recorded. A maximal incremental test in cycle ergometer (RacerMate CompuTrainer - Seattle, USA) was conducted to determine the following variables: a. lactate threshold (LT), b. peak oxygen consumption (VO2Peak), c. maximal heart rate (HRMax), and d. peak power (WPeak). The VO2Peak was used to divide participants into two groups: lower conditioned (LC—considering the 10 participants with lower VO2Peak) and higher conditioned (HC—considering the 10 participants with higher VO2Peak). In the second visit, participants performed the self-selected exercise session. During the self-selected exercise session, the participants were instructed to self-adjust the exercise, maintaining a continuous pattern. The instruction to perform the self-selected exercise was previously adopted in the literature [19]. During the exercise, no information was given to the participants (e.g., HR, power output, duration). Therefore, the exercise intensity and total duration were defined by each participant based on their perception. The power output (expressed as the percentage of WPeak obtained during the incremental test—%WPeak) and the percentage of VO2R (%VO2R) were recorded continuously during the self-selected exercise. The affective response was measured before and during the exercise session using the Feeling Scale (FS) [20]. All experimental procedures were performed between 13h and 17h. The ambient temperature and relative humidity during the tests ranged from 22 to 24 °C and 50–70%, respectively.

Participants

Twenty healthy male volunteers were invited to participate in the present study. The participants were divided into two groups (higher and lower conditioned) according to their VO2Peak. Therefore, ten participants with higher VO2Peak were included in the HC group, while the other ten participants with lower VO2Peak were included in the LC group. Based on the VO2Peak and age of participants (as presented in Table 1), it is possible to classify the cardiorespiratory fitness of the HC and LC groups respectively in the percentiles 40 (Fair) and 90 (Excellent) according to the ACSM [3]. The participation criteria were as follows: participants should be classified as “low risk” [3]; with no mental illness diagnosis or musculoskeletal injuries during the data collection; resting blood pressure <140/90 mmHg; aged between 18 and 45 years. In addition, all participants were asked to sign a consent form. All procedures performed in the study were in accordance with the 1964 Helsinki declaration, and this study was approved by the institutional ethics committee.

Table 1 Participants' characteristics
Full size table

Maximal incremental test

The participants were instructed to eat light food 2 hours before the test. No coffee, tea, or alcohol intake was allowed for 12 h before the experiment. Also, participants were asked to avoid exercise for at least 48 h before testing. An initial power output of 50 watts (W) was applied. After every 4-min stage, an increase of 30 W was conducted until voluntary exhaustion. This protocol was adopted to determine the lactate threshold [21]. Participants were asked to maintain a pedal cadence between 70 and 90 rpm (the cycle ergometer adjusts the pedal resistance to maintain the power output independently of the pedal cadence). The respiratory variables were recorded continuously during the maximal incremental test (Cortex Metalizer II, Cortex biophysik GmbH. Leipzig, Germany) to determine the VO2Peak (defined as the average of the last three records obtained during the test). Thirty seconds before the end of each stage of the maximal incremental test, a blood sample (≈ 25 μL) was collected from the ear lobe to determine the blood lactate concentration (Yellow Springs 1500 Sport, YSI Inc, Ohio, USA). The WPeak was defined as the last power output obtained during the test, and the lactate threshold was determined using the maximal distance method [22]. All procedures were performed with ambient temperature and relative humidity previously described.

Self-selected exercise

Participants were asked to perform a self-selected exercise session (adjusting the power output and the exercise duration). The following instruction was given to participants: “You should perform an exercise of a continuous nature. The intensity and duration of the exercise will be defined by you, and none of the training variables will be imposed on you in this exercise session”. This instruction was previously adopted in the literature [19]. During the exercise session, participants could change the pedal cadence as well as the front and rear gears. The exercise termination was defined individually by each participant.

Affective responses

The Feeling Scale (FS) was used to measure the affective response. The scale ranges from  – 5 (Very Bad) to +5 (Very Good) [20]. The FS was measured five minutes before the self-selected exercise session to verify if both groups presented the same affective response at the beginning of the exercise. During the self-selected exercise session, the FS was applied at 5-min intervals.

Physiological measures

The oxygen consumption (VO2) was recorded continuously during the self-selected exercise. A gas analyzer was used (Cortex Metalizer II, Cortex biophysik GmbH. Leipzig, Germany) to measure the respiratory variables. The VO2 mask and equipment were placed on the participant after they were positioned to perform the exercise. A face mask (Hans RudolphTM, Kansas, MO, USA) that covered the mouth and nose of the participant was attached to a bidirectional digital flow valve and fastened using a mesh hairnet and velcro straps. The calibration procedure was performed before each test according to the manufacturer's instructions.

Mechanical power

The power output performed in the cycle ergometer was recorded continuously during the exercise (RacerMate 3D Software, Seattle, USA). Then, the raw data of power output exported to a spreadsheet to be analyzed.

Statistical analysis

The average VO2R and exercise duration were compared to the VO2R and exercise duration recommended by the ACSM. An independent t-test was performed to compare the physiological (HRMax, LT, and VO2Peak), anthropometric characteristics, peak power (W.Kg-1), and total duration between groups (LC and HC). The power output (expressed as the percentage of peak power obtained during the incremental test – %WPeak) and the percentage of VO2R (%VO2R) recorded during the self-selected exercise were divided into percentiles (10% of the exercise) and compared using a repeated-measures ANOVA 2 × 10 (group vs. moment). Because the exercise duration was different for each group (LC and HC), the FS recorded during the exercise was divided into quartiles to reduce and equalize the data to four mean values (Q1, Q2, Q3, and Q4) and a repeated-measures ANOVA 2 × 5 [group vs. moment; pre and during (Q1, Q2, Q3, and Q4) was performed to compare the FS. For participants who performed the self-selected exercise session with less than 20 minutes (and consequently, with less than four FS data), preventing the use of quartiles for the analysis, average values between FS responses were used to complete the missing data. The level of significance was set at P ≤ 0.05. Inferential statistics were performed using the Statistical Package for the Social Sciences 23.0 (SPSS).

Results

Participants’ characteristics

The participants’ characteristics are presented in Table 1.

Self-selected aerobic exercise session

A significant difference was found between the total exercise duration performed by the LC (27.6 ± 11.0 min) and HC (44.5 ± 11.5 min) groups (t = − 3.32; p < 0.01; CI95% =  – 27.46 to  – 6.19) indicating that the HC group performed a higher exercise duration when compared to the LC group. In this respect, the descriptive analysis showed that two participants of the LC group performed an exercise duration lower than the recommended by the ACSM [3], and one participant of the HC group performed an exercise duration higher than the maximum recommended by the ACSM [3] (Fig. 1).

Fig. 1
figure 1

Average intensity (%VO2R) and duration performed by each participant

Full size image

The %WPeak presented a main effect for moment (F(2.79, 50.35) = 8.28; p < 0.01; ƞ2 = 0.31), indicating an increase of the power output during the exercise session in both groups. However, a main effect for group (F(1, 18) = 0.20; p > 0.05; ƞ2 = 0.01), or interaction (F(2.79, 50.35) = 1.79; p > 0.05; ƞ2 = 0.09) was not observed for %WPeak (Fig. 2). The %VO2R presented a main effect for moment (F(2.86, 51.52) = 17.94; p < 0.01; ƞ2 = 0.49), indicating an increase of VO2 throughout the exercise session for LC and HC groups. A main effect for group (F(1, 18) = 1.55; p > 0.05; ƞ2 = 0.07) or interaction (F(2.86, 51.52) = 1.21; p > 0.05; ƞ2 = 0.52) was not found. These data are presented in Fig. 2.

Fig. 2
figure 2

%WPeak and %VO2R adopted by both groups (HC and LC). LT-LC lactate threshold of the lower conditioned group, LT-HC lactate threshold of the higher conditioned group

Full size image

Affective responses

For the FS, we observed no main effect for group (F(1, 18) = 0.264; p > 0.05; ƞ2 = 0.01), moment (F(4, 72) = 0.911; p > 0.05; ƞ2 = 0.04) or interaction group vs moment (F(4, 72) =0.902; p > 0.05; ƞ2 = 0.04). These data are presented in Fig. 3.

Fig. 3
figure 3

Affective responses of both groups (HC and LC) before and during the exercise session

Full size image

Discussion

The objectives of the present study were to compare the intensity and duration selected by participants with higher and lower cardiorespiratory fitness to that recommended by the ACSM [3]. The results of the present study showed that most of the participants performed the self-selected exercise within the ACSM recommendation range, suggesting that self-selected exercise does not increase the risks on the individual's health (significantly, injuries related to high-intensity exercises) as well as reduce the expected benefits provided by the exercise when compared to imposed exercise. With respect to the exercise intensity, we observed that the LC group performed a % of WPeak (at 70% to 80% of exercise completion) above the lactate threshold. For the affective response, no difference was demonstrated between the groups, and the FS showed that the participants performed the self-selected exercise, maintaining a positive affect.

In a previous study, Johnson and Phipps [7] investigated self-selected exercise characteristics in 33 women performing exercise regularly for at least three months. The authors verified that only three women performed the self-selected exercise out of the range recommended by the ACSM [1]. In addition, the women performed the exercise session with an average duration of 45.1 minutes and an average rating of perceived exertion (RPE) of 13.9 a.u., according to Borg’s scale [23]. These values are within the limits suggested by the ACSM [1]. The authors inferred that the women performed intensity below the lactate/ventilatory threshold based on the total exercise duration (45.1 minutes). These results are similar to the present study and reinforce the hypothesis that the self-selected exercise does not increase the risk or reduce the benefits provided by the exercise. However, Johnson and Phipps [7] did not distinguish the participants according to the fitness level, contributing to these results. In the present study, we identified that participants with different exercise experiences might safely use the self-selection of exercise intensity.

Regarding the pacing strategy, no significant difference was found between the HC and LC groups for %WPeak. However, a significant main effect for the moment was found, indicating that the participants of both groups increased the exercise intensity throughout the exercise session (Fig. 2). Rose and Parfitt [8] showed that participants increased the exercise intensity during a self-selected exercise session. This fact may be related to the greater difficulty in performing the aerobic exercise at fixed-pace [6] compared to an exercise of the same average intensity performed at a self-selected pace. Although the similarities, in the present study, the LC group reached an intensity above the lactate threshold that did not occur with the HC group. However, this intensity was not maintained (Fig. 2). This result indicated that the HC group performed a pace adjustment only after the intensity had exceeded the lactate threshold, which may be due to the lower experience with exercise.

With respect to the exercise intensity, both groups maintained the power output close to the lactate threshold. This result is consistent with the Dual-Mode Theory, which is based on the premise that the lactate threshold is the physiological marker for the shift in the affective responses [24]. Despite the difficulty in its measurement, the lactate threshold should be considered in future aerobic exercise guidelines due to its influence on affective responses. The HC group performed exercise duration significantly higher than the LC group. However, most of the individuals finished the exercise session within the recommended range [3]. The exercise recommendations indicate that individuals with lower physical fitness should perform shorter exercise sessions [3, 5]. The results of the present study suggest that the individuals tend to finish the exercise session spontaneously within safe limits.

Although no significant differences were found for group or moment, the affective responses remained positive from pre-to-post exercise. In a review study, Rhodes and Kates [13] showed that changes in affective responses during exercise sessions of moderate intensities could predict exercise adherence. Based on this premise, the exercise session in the present study may induce exercise adherence; however, this hypothesis should be tested in long-term studies. To our knowledge, self-selection was not tested chronically. However, Lins-Filho et al. [25] showed that older women presented reliable psychophysiological responses between two self-selected sessions of 20 minutes. Therefore, it is plausible to hypothesize that self-selection may be chronically efficient. In addition to the information above, it is necessary to consider that exercise sessions of self-selected or imposed pace result in similar affective responses when performed at the same intensity [17]. However, it should be considered that self-selected exercise allows individuals to choose intensity and duration compatible with their optimal physiological and affective state [16, 17, 26].

The present study has limitations that should be considered. Only one exercise session was performed. Future long-term studies should be conducted to confirm these results. In addition, the present study was not performed with sedentary participants. Individuals with no aerobic exercise experience may adjust the exercise session with different characteristics observed in the present study.

Conclusion

The present study has practical implications for the exercise prescription, considering that most individuals self-selected their intensity within the ACSM’s recommended range [3]. Furthermore, only three participants performed exercise durations different from those recommended by the ACSM [3], suggesting that self-selected exercise sessions may be safely used for individuals with different cardiorespiratory fitness levels. Therefore, fitness professionals may include self-selection in the aerobic exercise program of different individuals to attain a stable positive affective response, as observed in the present study. However, future studies should include sedentary participants to test the applicability of self-selected aerobic exercise in this group of individuals. Also, the lactate/ventilatory threshold seems to be the physiological upper limit for the intensity self-selected by the participants in continuous aerobic exercise. In this sense, institutional statements should consider this variable in future aerobic exercise recommendations.