Resistance training (RT) is associated with numerous positive health outcomes such as improved cardiovascular health and reduced risk of all-cause mortality (1, 2). Hence, performing RT two or more days per week for the major muscle groups has been recommended for adults by both the American College of Sport Medicine and the World Health Organization (3, 4). However, only one out of three adults fulfill the weekly RT recommendations (5, 6) where lack of time is one of the most commonly reported barriers to not perform RT (7, 8). Consequently, time-efficient RT strategies has been investigated and suggested as an option to increase the compliance to RT programs (9).

Superset, i.e., performing two exercises after the other with no, or limited, rest between them (10), has been proposed as a time-efficient method for RT, as it roughly halves the training time compared to traditional RT (11, 12). Superset has shown to induce greater neuromuscular fatigue (13) and increased levels of metabolites such as blood lactate, creatin kinase concentration and testosterone compared to traditional RT session (14–16). Further, the increased level of metabolic stress could potentially influence the total training volume in a RT session which may potentially have an effect on muscle strength and hypertrophy (17, 18). Nonetheless, studies comparing volume between superset and traditional RT shows conflicting results (13, 19–22) with some studies report no differences in volume (21, 22), while others report a significantly higher volume for superset compared to traditional RT (13, 20). Importantly, several of these studies investigated its effects across only two exercises (13, 20–22) and the results may not be transferable to RT sessions including several exercises muscle groups. To the best of our knowledge, only one study has compared volume between superset and traditional in a whole-body RT session (19). Andersen et al. (19) compared a traditional and supersets session consisting of 8 exercises with 3 sets per exercise at an intensity of 9-RM in resistance trained adults. The superset session led to a decrease in training volume (measured as completed repetitions) by 4.2% compared to the traditional session. These findings indicate the importance of examining complete training sessions, as that superset's limiting effects on volume first becomes apparent when training volume is substantial.

How physical activity is perceived has shown to influence an individual's choice to continue with that activity later (23, 24). If compliance to RT is of importance, this could be of equal important as time-efficiency as it is unlikely to continue with something that is not pleasant or enjoyable. To the authors best knowledge, the previously mentioned study is also the only study that has compared the affective responses between superset and traditional RT when conducting a full-body workout. Andersen et al. (19) demonstrated a significantly higher rating of perceived effort and discomfort in the superset session, and a tendency for the superset session to be more pleasurable compared to traditional RT. There was no difference for enjoyment between the protocols (19). Furthermore, 62% of participants reported that they would prefer to continue with the superset RT if they had to choose (19). Notably, the study is limited by not measuring any physiological measures, e.g., heart rate or blood lactate. Adding such measurements could provide additional insight to the relationship of physiological measures and affective responses to superset RT. Furthermore, Andersen et al. (19) included RT individuals, and the affective responses may not be representative for other populations e.g., untrained individuals. Finally, the study did not report the affective measures during the sessions, hence it is unknown how or if the affective responses changed during the sessions. Therefore, the primary aim of this study was to compare the affective responses during and at the end of a whole-body superset vs. a traditional RT session in untrained individuals. Secondary, we compared physiological responses and total training volume (measured as total completed repetitions) to add additional insight into the acute effects of the different protocols.

We hypothesized that the superset RT session would lead to a higher perceived effort and discomfort and be perceived less pleasurable and less enjoyable than the traditional RT session both midway and post-exercise. Additionally, we hypothesized that the superset RT session would reduce the training volume (i.e., reduce total number of repetitions in session), reduce total time to conduct the session, induce higher blood lactate levels and a higher average heart rate compared to the traditional RT session both midway and post-exercise. Lastly, and based on the affective responses, we expected that most participants would prefer the traditional RT session.

There was a statistically significant interaction effect in RPE (p < 0.001) and discomfort RPD (p < 0.001), while there were no differences for sPDF (p = 0.404) or EES (p = 0.829) within or between the two experimental sessions. All details are presented in Figure 2.

Comparing effort (RPE) between sessions, post hoc corrections demonstrated a statistically significant higher effort in the superset RT at the middle (p = 0.012, r = 0.54) and post-exercise (p = 0.016, r = 0.53) compared to traditional RT. For the traditional RT session, the effort increased significantly from the middle to post-exercise (p = 0.028, r = 0.49). No difference between middle and post-exercise were observed for the superset RT session (p = 0.368, r = 0.31).

For discomfort (RPD), no statistically significant difference was observed between the sessions in the middle (p = 0.336, r = 0.32) or at the end of the sessions (p = 0.092, r = 0.41). However, a statistically significant difference was demonstrated from the midway to post-exercise of the superset session (p < 0.001, r = 0.62), but not within the traditional session (p = 0.128, r = 0.39).

For blood lactate, there was an interaction effect between time and modality (F = 11.432, p < 0.001, Table 2). The post hoc analyses showed no difference between the modalities before the start of the sessions (p = 0.370, d = 0.17, Table 2). However, in the middle of the session (12.9%, p < 0.001, d = 0.73) and at the end of the sessions (18.3% p < 0.001, d = 0.95), the superset session led to higher lactate values than the traditional session. When analyzing the lactate levels within each modality we observed the same pattern for both modalities. Both sessions led to higher lactate levels at middle (Traditional; 262.5%, p < 0.001, d = 2.56, Superset; 296.6%, p < 0.001, d = 3.47) and end of the session (Traditional; 240.6% p < 0.001, d = 2.59, Superset; 290.9% p < 0.001, d = 3.74), compared to before the start of the session. There were no differences between the middle and the end of session for any of the modalities (Traditional; p = 0.265, d = 0.32, Superset; p = 1.000, d = 0.12).

The traditional session had a mean heart rate of 128 ± 16 beat per minute (bpm), while the superset session had a mean heart rate of 138 ± 16 bpm. The mean heart rate in the superset session was significantly higher compared to the traditional session (9.9 ± 7.2 bpm, p < 0.001, d = 1.53).

Comparing the total repetitions between sessions, the traditional session had 4.6% more repetitions compared to the superset session (158 ± 18 vs. 151 ± 17 repetitions, p = 0.006, d = 0.54, Table 3), and took 60% longer time to complete (40 ± 2 vs. 25 ± 2 min, p < 0.001, d = 6.62).

Finally, twenty out of the thirty participants reported that they would prefer to continue with superset training if having to choose between the two.

Consistent with our hypothesis and other studies (14–16, 19), the perceived effort in the superset session in the present study was higher compared to the traditional session. The higher perceived effort in the superset session could be explained by the more intense workout, i.e., same work in a shorter period of time. This would increase the metabolic stress which has been coupled with increased rating of perceived effort (16). This speculation corresponds with our observations of increased levels of blood lactate and average heart rate for the session. Higher levels of blood lactate have been associated with higher perception of effort (15, 16), while heart rate has shown strong relationship with the perception of effort in multiple sports (35). These higher levels of metabolic stress could also explain the tendency for the increased perception of discomfort for the superset session. Higher levels of metabolic stress have shown to increase the perception of fatigue and pain/discomfort (36). Furthermore, it is reported that the perceived effort and discomfort have a significant, but weak correlation to each other (30). Therefore, the increase in both could partly be explained by each other.

Notably, the increased perception of effort and discomfort did not lead to a change in perception of pleasure/displeasure or enjoyment between the sessions. These findings were in contrasts with previous studies indicating that the perception of effort has a negative association with the feelings of pleasure/displeasure scale (32, 37, 38). For example, Almeida et al. (38) compared traditional resistance training with more intense modalities using shorter rest intervals (rest-pause training) or shorter rest intervals and reducing intensity (sarcoplasmatic stimulating training) in male bodybuilders. The results showed an increased perception of effort and discomfort, and reduced pleasure for the two more intense modalities compared to traditional training. Importantly, the differences in study design and training experience makes it difficult to compare the findings to our results. More similar to our design, Andersen et al. (19) compared the perception of pleasure/displeasure (sPDF) and enjoyment (EES) between superset and traditional RT in a whole-body RT session and demonstrated a tendency for the superset session to be perceived as more pleasurable (p = 0.059) while there was no change for enjoyment (p = 0.661). Of note, in contrast to our population, the population in Andersen et al. (19) had a mean average resistance training experience of 8.4 (±6.6) years, which may indicate they are more used to and find more pleasure in the more intense superset RT.

There were no changes in the perception of pleasure/displeasure or enjoyment from the middle to post-sessions. In contrast, most people (>95%) experience an immediate positive response (i.e., affective rebound effect) after cessation of aerobic exercise (39). It has been suggested that the affective rebound effect becomes evident after exercise leading to an affective decline. In our study the rating of perceived enjoyment and pleasure were generally high during the sessions, which arguably could reduce the potential of a rebound effect. These explanations are consistent with previous studies suggesting that the affective rebound effect is less pronounced in resistance training exercises (40, 41).

In the present study, a 4.4% decrease in total completed repetitions were observed in the superset session compared to traditional session when performing sets to voluntary failure. This is a similar decrease in repetitions (4.2%) as the participants in Andersen et al. (19). Of note, previous studies have reported similar volume (15, 42) or higher volume (13, 20) conducting superset compared to traditional RT. However, these studies are limited by only examining two exercises (13, 15, 20, 42). The findings in our study supports the notion that the decrease in volume first becomes apparent in RT sessions when including several exercises. This is strengthened when comparing the repetitions between the first two exercises in our study, resulting in no difference between the protocols (difference: 1.8 repetitions, p = 0.119).

The reduction of completed repetitions, i.e., training volume, found in our study could be explained by the increased metabolic stress (i.e., increased blood lactate and heart rate) as a result of reduced time to rest. Less time to rest would reduce the time to restitute between the sets/exercises and increase the metabolic and neuromuscular stress (43, 44). An increase in metabolic stress has been associated with a lower performance in volume due to reduced capacity to sustain muscular force (17). Our finding are comparable to several studies reporting an increased metabolic stress in superset compared to traditional RT (15, 16, 45). Although speculative, over findings may also be a result of increased central fatigue during the superset session (44).

The present study has several limitations which needs to be addressed. First, the participants in the study were unfamiliar with the scales and RT before the start of this project. The lack of RT experience may have affected their responses on the affective outcomes due to pre-assumed expectation, lack of sensation, or stable base-line references of the two RT sessions. Importantly, all participants were familiarized with the questions and scales before the experimental sessions. Furthermore, the sessions were randomized, and the comparisons were within participants possibly nulling out any familiarization effect. Secondly, the study aimed to target approximately 10-RM in each exercise. However, some of the participants in the experimental session managed to lift more repetitions (e.g., 13 and not 10 repetitions on average in the first exercise). We would preferred to have a more similar number of repetitions between the exercises, however and importantly, the sets in both modalities were conducted with the same loading to failure, and the order of the sessions were randomized (i.e., a systematic error affecting both RT sessions). Also, even if the number of repetitions/intensity to some degree extended 10-RM, it could still be argued to be well within an effective intensity for increasing muscle strength, as long as the sets were performed to failure (46). Furthermore, the short-term learning effects between familiarization- and the experimental sessions was potentially reduced by using training machines and not free-weights (47). Still, future studies should investigate individuals’ perceptions to superset across different loading and sets in addition to examine the impact on affective responses over time conducting the two RT session among untrained participants. Lastly, the participants in this study were untrained adults and the findings can therefore not be generalized to other populations.

Interestingly, 20 out of 30 participants choose superset as their preferred sessions if they were to continue RT. The main reasons given for this choice was time-efficiency and a feeling of a harder/more effectful workout. Among the 10 participants preferring traditional RT, the main reasons were a feeling of being able to perform better with the traditional RT and the superset RT was perceived as too strenuous to continue over time. Considering that superset leads to greater metabolic stress (i.e., blood lactate and heart rate) and higher perceived effort, it is of interest how untrained individuals would adhere to superset RT over time. Therefore, from a public health perspective, it would be of interest to conduct longitudinal studies to investigate the efficacy and adherence of prescribing superset as RT modality to untrained adults in real-world settings (i.e., without supervision sessions).

In conclusion, one single session of superset RT was perceived as more effortful and led to higher levels of blood lactate, increased average heart rate, a decrease in volume and took less time compared to a single session of traditional RT in untrained adults. There were no statistically significant differences between the sessions for discomfort, session pleasure/displeasure or enjoyment, however, 20 out of the 30 participants preferred the superset session.