An analysis of the perceived causes leading to task-failure in resistance-exercises

Introduction

Whether one should reach task-failure in resistance-exercises is a question that attracts scientific and applied interest (Nóbrega & Libardi, 2016; Davies et al., 2016; Sampson & Groeller, 2016). Here we refer to task-failure as either momentary-failure (MF), the point in which an attempted repetition cannot be completed with proper form, or repetition maximum (RM), the final repetition one can complete prior to reaching MF (Steele et al., 2017). It has been established that reaching or approaching task-failure is important for hypertrophy and strength development (Davies et al., 2016; Morton et al., 2016). Yet, to date, little is known about the subjective reasons underpinning task-failure. Although it can be expected that the inability to generate the required force with the target muscles is the main reason for task-failure, perception of fatigue, negative feelings, cardiovascular strain and pain can also be at play. Reaching task-failure due to one or more of the aforementioned factors could lead to different adaptations. For example, reaching task-failure because one cannot generate enough force with the target muscles is possibly more effective for hypertrophy purposes compared to reaching task-failure because of cardiovascular strain, or pain in body parts other than the target muscle groups. While investigating the underlying physiological causes of task-failure is a challenging task, examining the subjective aspects believed to cause task-failure is attainable and can shed light on this important issue.

One effective way to study subjective experiences during resistance training is with single-item scales (Buckley & Borg, 2011; Helms et al., 2016; Hackett et al., 2017). For example, rating of perceived effort (RPE) scales, such as Borg CR-10 (e.g., Buckley & Borg, 2011), can assist quantifying how much effort one is investing during, or after set completion, and allow for comparisons between different exercises, loads, and body parts involved in the exercise (Duncan, Al-Nakeeb & Scurr, 2006; Buckley & Borg, 2011). Interestingly, when sets are taken to task-failure, perceived effort is not always rated as maximal, indicating that effort is not the limiting factor in such cases (Pritchett et al., 2009). Such results led investigators to examine if other constructs can be the reason people terminate sets (Steele et al., 2016). A number of studies found that perception of discomfort, rather than RPE, was maximal at the point of set-termination, suggesting that discomfort can be a limiting factor (Fisher & Steele, 2017; Stuart et al., 2018).

Other popular effort scales gauge how many repetitions trainees estimate they can complete before reaching task-failure (Helms et al., 2016; Hackett et al., 2017). To illustrate, a rating of two indicates that only two more repetitions are left before reaching task-failure (Hackett et al., 2017). These scales are similar to RPE scales although developed solely for resistance training (Helms et al., 2016; Hackett et al., 2017). While single-item scales are practically useful and lead to theoretical insights (Halperin & Emanuel, 2019), they lack the resolution required to pinpoint the reasons why task-failure occurs. This is because they depend on a single question, which revolves around a single construct, to which people can only provide a single answer. Hence, other strategies of investigations are warranted.

Another way to investigate the causes leading to task-failure is to ask subjects why they terminated a set, and allow them to answer in an unrestrictive manner. This measurement strategy can expand our understanding of set termination that goes beyond the insightful, yet limited knowledge gathered via single-item scales. Such knowledge could lead to new research avenues. The limiting factors may vary as a function of the exercises completed and the loads lifted. Hence, the purpose of the present work was to examine the perceived causes trainees attribute to task-failure across different loads and exercises. To achieve this goal, we instructed resistance-trained subjects to reach task-failure in the squat and bench-press exercises under two load conditions (70% and 83% of 1RM). Within 20-40 s after set completion, we asked them what was the limiting factor in the set, and why they did not perform another repetition. Subjects’ answers were recorded, transcribed and analyzed.

Materials & Methods

Results

Overall, 158 statements from terminated sets were recorded with only 37 sets terminated due to MF and the rest due to RM. We had ten missing observations in the 83%1RM bench-press condition; four due to drop-outs and six due to technical difficulties. We had four more missing observations across the other conditions and exercises due to technical difficulties. The mean number of repetitions completed across sets and gender in the 70%1RM condition was 16.4 ± 8.9 (range: 6–55) and 13 ± 2.9 (range: 7–19) in the squat and bench-press respectively, whereas in the 83%1RM it was 8.1 ± 3.3 (range: 3–17) and 7.1 ± 1.8 (range: 4–12) in the squat and bench-press respectively.

Overall frequencies of each category. The final mixed logistic regression model included a random intercept, and its fixed effects were defined by the following equation:

Frequency_ij ∼logistic(Category _ij+ u_0j)

Out of the 158 statements, muscle fatigue was found to be the most frequent category accounting for 53.8% of the terminated sets (OR = 9, b = 2.2, SE = 0.29, z = 7.42, p < 0.001, 95% CI [1.8–2.9]). The second most frequent category was general fatigue accounting for 25.9% of the terminated sets (OR = 2.7, b = 1, SE = 0.31, z = 3.24, p = 0.001, 95% CI [0.45–1.67]), followed by pain (12%), cardiovascular (11.4%) and negative affect (10.1%). No significant differences were found between the frequency of either pain (OR = 1, b = 0.06, SE = 0.35, z = 0.17, p = 0.861, 95%CI [−0.66–0.74]) or negative affect (OR = 0.87, b =  − 0.13, SE = 0.36, z =  − 0.36, p = 0.716, 95%CI [−0.87–0.59]) and the cardiovascular category (See Fig. 1).

Differences in frequencies between exercises and loads, males and females. The results of the mixed logistic regression models tested for each separate category are presented at Table 3. The frequencies the categories by gender are reported in Table 4.

Discussion

Here we examined what limiting factors were perceived to lead to task-failure in the squat and bench-press exercises using 70% and 83% of 1RM, among resistance-trained subjects. Subjects were required to state what were the limiting factors of each set within 20–40 s after reaching task-failure. Of the five categories used to map subjects’ responses, specific muscle fatigue was the most frequently reported. Within the muscle fatigue category, the target muscles (i.e., prime movers) in each exercise were specified in most, but not all of the responses. The remaining responses were categorized as general fatigue, pain, negative affect and cardiovascular strain. Some differences were observed between exercises with task-failure in the squat being attributed more often to negative affect and, while statistically insignificant, in cardiovascular strain compared to the bench-press exercise. Similarity, some differences were observed between loads, in which lighter loads led to greater pain and cardiovascular strain compared to heavier loads. No differences in the distribution of categories were identified between males and females.

As can be expected, most sets were perceived to be terminated because of muscle fatigue attributed to target muscle groups (i.e., lower body for squats and upper body for bench-press). However, within this category, certain variability was noted. In the squat, 27% (13/47) of the statements within the muscle fatigue category were attributed to the upper body and lower back, indicating that in these cases, task-failure was not perceived to be due to the target muscles. Moreover, different muscles within the legs were mentioned, including the gluteus, quadriceps and hamstring, indicating that limiting factors within the target muscle groups can vary between individuals. In the bench-press, some variability was also noted within the target muscle groups, with the chest muscles being reported the most, followed by the arms (presumed to be triceps) and then the shoulders. The variability in reporting which of the target muscle groups was the limiting factor in each exercise can partly be explained by technical variation in the way the exercises were performed. For example, different stance widths, foot placement, and forward lean during the squat can affect the moments at the hip and knee joints (Escamilla et al., 2001; Glassbrook et al., 2017; Lorenzetti et al., 2018). It is interesting to consider whether variations in the perceived muscle group being the limiting factor in a given exercise can lead to different adaptations, such as strength and hypertrophy. Alternatively, whether suboptimal adaptations may occur when unrelated muscles are considered to be the limiting factor in exercise-performance.

General fatigue was also frequently attributed to set termination across exercise and loads. This could indicate that multi-joint exercises involving large muscle groups produce a global and unspecified feeling of fatigue strong enough to lead to set-termination. While pain, negative affect, and cardiovascular strain were mentioned fewer times compared to specific and general fatigue, some interesting patterns emerged. Mainly, the results indicate that more sets in the squats were terminated due to pain and negative affect compared to bench-presses, and that in the 70%1RM squat condition more sets were terminated due to cardiovascular strain. Some of these findings are aligned with other studies. For example, lower body exercises have shown to cause greater degree of negative affect (Portugal et al., 2015), and require greater energy expenditure (Lyons et al., 2007; Flanagan et al., 2014; Vianna et al., 2014) compared to upper body exercises. These findings suggest that squats may be limited by a wider range of factors, compared to the bench-press. As a whole, these results are also alighted with other studies reporting that other constructs, such as discomfort, can be a limiting factor during sets (Steele et al., 2016).

This study has several methodological aspects worthy of discussion. First, asking subjects to answer the question concerning set termination within ∼30 s can be viewed both as a strength and a limitation. Asking the question in proximity to set termination was expected to lead to a more accurate answer, but given the physically challenging nature of the task, subjects may have provided less details. Similarly, subjects were allowed to answer the question without any restrictions, but this may have caused important information to be lost that a more structured questioning procedures might have captured. Second, only 23% of sets were completed due to TF, despite subjects being encouraged to reach TF in all sets. It is possible that set termination due to RM and TF lead to different perceptions of its underlying causes. However, since it is ethically impossible to enforce TF, overcoming this limitation is a challenging task. Third, while subjects in this study were experienced in RT, their unique training background could have shaped their responses. Moreover, subjects’ beliefs and attitudes about the factors leading to set termination, as well as their general knowledge pertaining to resistance training may have also played a role in their responses (e.g., Vaegter et al., 2020). Third, no physiological outcome, such as heartrate, was measured and correlated with participants’ statements. In view of the aforementioned points, future studies should collect more data on trainees’ training background and knowledge, attitudes and perception using structured or semi-structured interviews and also collect physiological measures.

As studies continue to investigate how many repetitions one completes relative to task-failure, it may be of added value to examine what are the limiting factors leading to set-termination. This is because different limiting factors within sets could lead to different acute and long-term adaptations that might not be aligned with the sought-after outcomes. Investigating limiting factors can be done by directly measuring the possible physiological pathways leading to set-termination. Such studies require complex designs and equipment. An alternative is to study the subjective experiences accounting for set-termination, as was done in the present study, and to our knowledge, for the first time. The present investigation may indicate that relying on task-failure as a standard for load and repetition prescription may fail to capture a variety of limiting factors other than fatigue of the target muscles. Thus, it could be that in addition to prescribing one to reach task-failure, or proximity to task-failure, monitoring other aspects of set-termination may prove beneficial. For example, loads can be modified in cases that cardiovascular strain or negative affect are reported at set-termination, and exercise technique can be modified in cases that target muscles are not perceived to be a limiting factor. However, it is required to first establish what type of relationships, if any, exist between perception of the limiting factors and actual adaptations.

Conclusions

We observed that set-termination was mostly perceived by subjects to be a result of muscle fatigue in the target muscles, followed by a general feeling of fatigue, negative affect, pain and cardiovascular causes. These reasons were found to vary between exercises and loads. The results of this study show that there are a variety of perceived reasons for set termination, which might affect exercise-adaptations, and warrant further investigation.

Supplemental Information