Effects of a 12-Week Low-Intensity Resistance Training Program on Force-Matching Task and Balance in Young Men

Featured Application

Specialists promoting physical activity as a means of prevention of many diseases, including lifestyle diseases, and treating physical activity as an important element of public health, gain additional knowledge about the effects of resistance training. The systematic use of low-intensity resistance exercise, in addition to the well-known benefits, leads to improved accuracy in matching force to the demands of a motor task.

Abstract

Background: The effects of low-intensity resistance training on health and muscular performance have been widely reported, but its effects on motor skills such as balance and force matching have been overlooked. Hence, the purpose of this study was to determine the effects of low-intensity resistance training on a force-matching task and balance. Methods: The subjects from the intervention group (EXP; n = 20) participated in a 12-week low-intensity resistance training program. The measurements of balance and force-matching ability were conducted before and after the intervention. To determine the accuracy and steadiness (variability) in the force matching task, we calculated the values of three errors: (1) absolute error (AE), (2) constant error (CE), and (3) variable error (VE). Results: In the force-matching task performed after the training, the values of two errors decreased: (1) AE (right leg, p = 0.0008; left leg, p = 0.0008), and (2) CE (right leg, p = 0.0064; left leg, p = 0.0440). Resistance training did not significantly affect VE and the parameters characterizing COP sway in the balance test. Conclusions: The 12-week low-intensity resistance training improved the accuracy of the force-matching task but did not change postural stability or postural strategies.

1. Introduction

Resistance training has long been used by athletes to improve muscle strength, power, and hypertrophy. Since its beneficial effects on the human body are far broader, it has found applications in rehabilitation, disease prevention, and public health promotion. Resistance training has been shown to reduce body fat, increase basal metabolic rate, improve blood lipid profiles, enhance glucose tolerance and insulin sensitivity, and maintain long-term independence and functional capacity [1]. Untrained individuals should initially perform low-load exercises (>15 repetitions maximum (RM)) to avoid possible overloading of the musculoskeletal system and be able to focus attention on the correct technique while performing the exercises [2,3]. In addition to its health-promoting effects, low-intensity resistance training has been shown to improve local muscular endurance, power, dynamic strength (in previously untrained individuals), and flexibility (in individuals with poor flexibility) [1,3]. However, there are few reports on the effects of strength training on other manifestations of motor skills, such as balance and force control.

Precise force matching determines motor performance both in sports and everyday activities (e.g., grasping a plastic cup with a drink) [4,5]. Further, inadequate force application ability enhances the risk of injury [6,7]. Muscle force perception is based on information from two sources. The first is formed by the Golgi tendon organs, which provide information about muscle tension. The second is formed by the collaterals of cortical motoneurons [8].

Studies on the impact of resistance training on force-matching ability mainly included patients with impaired proprioception, and their results are inconclusive [9,10]. Studies in an older population have confirmed the positive effects of resistance training on force accuracy and steadiness [9]. In contrast, studies involving patients with ankle instability showed no improvement in force application accuracy after a strength training program [10]. To date, there is a lack of research on this topic involving young and healthy individuals, except for one study that reported that 12-week high-intensity strength training enhanced the accuracy and steadiness in the force matching task in young, healthy men [11].

Balance is based on multiple components that coordinate sensory input and motor output [12]. In young adults, impaired postural control is an important risk factor for falls and injuries related to physical activity [13]. Research conducted on young basketball players showed that high variations of postural sway in both anteroposterior and mediolateral directions during one-legged stance were associated with the occurrence of ankle injuries [14]. As with force matching, there is also a lack of studies on the effects of resistance training on balance in healthy adults. This should be explained by the fact that the risk of falls and injuries due to deterioration of postural control increases with age and as a result of various diseases. Therefore, in elderly and patient populations, researchers are looking for effective methods to prevent or improve the deterioration of balance control. Systematic reviews and meta-analyses confirm that resistance exercise is an effective method for improving balance in older people [15,16]. The improved balance may be due, among other things, to the fact that resistance exercise attenuates muscle sarcopenia, which occurs as a result of aging [1].

We can improve force-matching accuracy and balance with dedicated exercises, but implementing them into a training program involves additional time we have to devote to training and sometimes the use of additional equipment or devices. Therefore, in our study, we decided to investigate whether widely used low-intensity resistance training would also positively affect such motor skills as force matching and balance. The previous study [11] confirmed that high-intensity strength training improved force-matching ability in young, healthy men. Although the mechanisms leading to this could not be elucidated, it seems likely that the improvement in force matching was the result of systematic resistance exercise over a long period (12 weeks) rather than the effect of using heavy loads. Although force matching and postural control appear to rely on somewhat different proprioceptive senses, they may be related in some way, and any similarities or discrepancies may shed more light on the mechanisms under study [17]. Hence, the purpose of this study was to determine the effects of low-intensity resistance training on the force-matching task and balance. We hypothesized that low-intensity resistance training would improve force matching and balance in healthy young men.

2. Materials and Methods

2.1. Study Design

This study design was a randomized, controlled trial with repeated measures [18]. The allocation of groups was randomly generated using Research Randomizer v.4. Participants were assigned into two groups: (1) an intervention group (EXP), and (2) a control group (CON). Force matching and balance were measured one week before (Pre) and one week after (Post) the intervention. Force accuracy and steadiness were measured by pressing the foot on the force plate with a predetermined force. We calculated absolute error (AE) and constant error (CE) as measures of force accuracy, and variable error (VE) as a measure of steadiness. Balance was assessed on a Kistler force platform. During each test (Pre and Post), two trials of normal bipedal quiet stance were performed with eyes open: one on a firm surface, and one on a foam surface. The trial order was counterbalanced.

2.2. Participants

Forty men (age = 21.5 ± 1.9 years) gave their informed consent to the experimental procedure. Inclusion criteria: male; age 19–24 years; and active lifestyle (1.5 h of physical activity at least 3 times a week). Exclusion criteria: professional athlete; musculoskeletal disorders; any contraindication to resistance training; balance disorders; and absence for more than 6 training units. The participants were randomly assigned to the intervention (EXP, n = 20) and control (CON, n = 20) groups. The groups did not differ significantly in terms of age and anthropometric parameters (

Table 1. Participants’ characteristics; mean ± SD.

	EXP (n = 20)	CON (n = 20)	p-Value
Age (years)	22.0 ± 1.7	21.1 ± 1.8	0.304
Body mass (kg)	76.9 ± 7.2	73.0 ± 8.7	0.292
Height (cm)	179.9 ± 6.0	176.9 ± 7.8	0.367
BMI	23.7 ± 1.6	23.3 ± 2.0	0.777

).

The study was approved by the local ethics committee (Ethics Committee of Wroclaw University of Health and Sport Sciences; #8/2022) in accordance with the Declaration of Helsinki.

2.3. Resistance Training Program

One week before the intervention, a qualified coach instructed the subjects from EXP on the technique of each exercise. The next day, the twenty-repetition maximum (20 RM) was measured for each exercise. Throughout the study, 20 RM was re-measured every 2 weeks to adjust loads. The training sessions were held three times a week (Monday/Wednesday/Friday) at the same time of day for 12 weeks. Participants performed a low-intensity circuit training program that consisted of 15 exercises that engaged the muscles of the entire body: underhand cable pulldowns, seated rows, pectoral machines, push-ups, shoulder presses, dumbbell lateral shoulder raises, horizontal leg presses, leg extensions, deadlifts, dumbbell lunges, dumbbell lateral lunges, kettlebell swings, burpees, crunches, and side crunches. This type of training program is used to adapt beginners to strength training and develop local muscular endurance. Participants performed 20 reps of each exercise with a 30 s rest between exercises to move safely from one exercise to the next. A total of 3 sets were completed, with a 3 min rest between sets. None of the exercises were aimed at improving the balance and accuracy of the force application.

Participants from CON performed recreational physical activity for about 90 min three times a week for 12 weeks.

2.4. Measurement of Force Accuracy and Steadiness

The equipment used to measure force accuracy is shown in Figure 1. A detailed technical description of the device can be found in our previous work [11]. Measurements were taken in a sitting position with the back resting on a backrest. The tested leg was placed on a pad attached to the board, while the untested leg rested on the ground. The seat was moved linearly to obtain a knee flexion of 60 degrees. To measure maximum isometric voluntary contraction (MIVC), the subject pressed the pad with maximum force and maintained pressure for 3 s. Three such trials were performed for the right and left legs, and the mean force value was calculated for each trial. The highest mean value (for each leg) served as a reference for determining the target force (50% MIVC) that the subject was expected to reproduce.

The force-matching task was to press the pad at 50% MIVC for 3 s. The learning phase (5 reps for each lower limb) was conducted with visual control to facilitate force matching (50% MIVC). The actual measurement was made without visual feedback. Subjects pressed the pad five times for 3 s with each leg (in randomized order), trying to match the target force. We used a 10 s intra-limb rest and a 30 s inter-limb rest. The force signal was sampled at 100 Hz. The software calculated the mean force value for each 3 s trial.

The force matching task was conducted one week before (Pre) and one week after the intervention (Post), in the morning and without a warm-up. Using the following formulas, we calculated the values of three force errors: (1) absolute error (AE), (2) constant error (CE), and (3) variable error (VE).

$$\mathrm{A}\mathrm{E}=\left|\frac{\sum \left(x_i-T\right)}{n}\right|$$

(1)

$$\mathrm{C}\mathrm{E}=\frac{\sum \left(x_i-T\right)}{n}$$

(2)

$$\mathrm{V}\mathrm{E}=\sqrt{\frac{{\sum \left(x_i-\overline{x}\right)}^2}{n}}$$

(3)

where x_i is the force reproduced in trial i, $\overline{x}$ is the mean value of force from five trials, T is the target force (50% MIVC [N]), and n is the number of trials.

The AE determines the accuracy in the force matching task, the VE determines the steadiness (variability), and CE indicates a tendency to apply the force above (positive values) or below the target (negative values) [19].

2.5. Measurement of Balance

A force platform (Kistler 9281CA, Winterthur, Switzerland) was used to assess postural control. The subjects from both groups participated in two stabilographic tests: the first before (Pre) and the second after a 12-week resistance training program carried out in one of the groups (Post). During each test, two trials of normal bipedal quiet stance were performed with eyes open: one on a firm surface, and one on a foam surface. The trial order was counterbalanced. The foot position (5 cm apart) was standardized to ensure repeatability across trials and participants. During the data acquisition, subjects were asked to stand comfortably but as still as possible. The subjects were barefoot, with their arms relaxed at their sides. They were also asked to fix their gaze on a 5 cm X mark placed in front of them on a white wall at a distance of 3 m at eye level. One minute of rest was provided between the trials in order to minimize the effects of fatigue. The trials lasted 30 s with a sampling frequency of 100 Hz. The researcher pressed the start measurement button after receiving a “ready” signal from the participant. The measurement started 10 s later to avoid transient effects related to the subjects’ adaptation to the test being performed.

The recorded ground reaction forces were used to compute the center-of-pressure (COP) time series, which consisted of 3000 data points. The following sway parameters were computed in the medial-lateral (M/L) and anterior-posterior (A/P) planes: variability, mean speed, median frequency, sample entropy, and fractal dimension. The COP variability was measured by the time-series standard deviation, while the mean speed was computed as the recorded path length divided by the trial duration. Sample entropy and fractal dimension were computed based on Bieć et al. [12].

2.6. Statistical Analysis

Data were analyzed using Statistica 13.1 (Dell, Round Rock, TX, USA). Their compliance with the normal distribution was checked using the Shapiro–Wilk test. Some results of the force-matching task deviated from the normal distribution and required the use of non-parametric tests. Therefore, the Wilcoxon test was used for intragroup (Pre vs. Post) comparisons and the Mann–Whitney test for intergroup (EXP vs. CON) comparisons. Effect size was calculated according to the formula: ES = Z/√N, where a large effect is 0.5, a medium effect is 0.3, and a small effect is 0.1 [20].

In contrast, all parameters of the COP time series met the criteria of normal distribution. In order to evaluate the possible effects and interactions of resistance training (two groups), time (Pre, Post), and the support surface (firm, foam), a mixed analysis of variance was conducted on all sway parameters in each plane separately. Post hoc comparisons were performed using the Tukey test. The level of significance was set at α < 0.05.

3. Results

After 12 weeks of the experiment, a significant increase in MIVC was observed only in the EXP group (17.4% for the right leg and 17.1% for the left leg).

3.1. Force Accuracy and Steadiness

These results are presented in Figure 2. At the start of the study, there were no differences in absolute and constant errors in force-matching tasks between the two legs. Only the variable error was lower (Z = 2.88; p = 0.0040; ES = 0.64) for the left leg in CON than in EXP. Resistance training decreased the absolute error in EXP in the right leg (Z = 3.36; p = 0.0008; ES = 0.75) and in the left leg (Z = 3.22; p = 0.0008; ES = 0.72). The final values of this error were lower in EXP than in CON: (Z = 3.83; p = 0.0001; ES = 0.86) and (Z = 3.12; p = 0.0018; ES = 0.70) for the right and left leg, respectively. Similarly, the constant error improved (decreased) its values in EXP due to training in the right leg (Z = 2.73; p = 0.0064; ES = 0.61) and the left leg (Z = 2.02; p = 0.0440; ES = 0.45). The final values of this error were also lower in the EXP than in CON: (Z = 2.77; p = 0.0056; ES = 0.62) and (Z = 3.45; p = 0.0006; ES = 0.77) for the right and left leg, respectively. Only the variable error did not change its values in both groups after 12 weeks from baseline.

3.2. Balance

There were no main effects of group on sway indices, nor group x time interactions except a single group x time interaction (F (1.38) = 4.18, p = 0.0480) on the COP amplitude in the A/P plane, which was caused by the detrimental effect of time (increase in COP amplitude) on postural steadiness in the CON group on a foam surface only (p = 0.0450).

Several COP parameters exhibited the main effects of the support surface. The transfer from hard to foam surface increased COP variability in the M/L (F (1.38) = 34.49, p < 0.0001) and A/P (F (1.38) = 11.58, p = 0.0016) planes, and increased COP mean speed in the M/L (F (1.38) = 24.53, p < 0.0001) and A/P (F (1.38) = 24.73, p < 0.0001) planes as compared to standing on a firm surface. The same change in support surface decreased COP entropy (F (1.38) = 8.84, p = 0.0051), COP frequency (F (1.38) = 6.64, p = 0.0140), and COP fractality (F (1.38) = 4.06, p < 0.0512) in the A/P plane. The detailed results are presented in

Table 2. Mean values (SD) of the COP parameters for two groups (EXP and CON) standing on firm or a foam surface before (Pre) and after (Post) a 12-week exercise period in the frontal (M/L) and sagittal (A/P) planes.

	EXP				CON
	Pre		Post		Pre		Post
	Firm	Foam	Firm	Foam	Firm	Foam	Firm	Foam
Amplitude M/L (mm)	5.18 (0.85)	6.03 (1.23)	4.98 (0.78)	5.62 (0.97)	5.42 (1.20)	6.12 (1.05)	5.33 (0.82)	6.35 (1.93)
Amplitude A/P (mm)	6.61 (2.41)	8.55 (2.86)	6.95 (2.41)	7.86 (2.54)	7.22 (2.38)	7.81 (1.71	8.55 (2.90)	9.22 (3.25)
Mean Speed M/L (mm/s)	26.8 (5.8)	33.5 (10.8)	24.7 (4.9)	29.4 (6.7)	28.1 (8.3)	31.4 (9.2)	26.7 (8.9)	29.6 (11.9)
Mean Speed A/P (mm/s)	25.9 (5.4)	29.4 (8.3)	24.5 (5.6)	28.2 (6.6)	27.4 (8.6)	31.5 (8.3)	24.5 (7.1)	28.9 (12.0)
Frequency M/L (Hz)	1.03 (0.16)	0.97 (0.22)	0.94 (0.17)	0.92 (0.28)	0.95 (0.19)	0.89 (0.21)	0.98 (0.18)	0.85 (0.22)
Frequency A/P (Hz)	0.81 (0.18)	0.72 (0.16)	0.77 (0.17)	0.70 (0.11)	0.82 (0.19)	0.74 (0.16)	0.66 * (0.17)	0.69 (0.13)
Entropy M/L ( )	0.61 (0.07)	0.62 0 (0.09)	0.61 (0.06)	0.62 (0.09)	0.59 (0.08)	0.61 (0.09)	0.61 (0.09)	0.60 (0.09)
Entropy A/P ( )	0.78 (0.16)	0.67 (0.13)	0.71 (0.15)	0.67 (0.12)	0.73 (0.19)	0.70 (0.13)	0.63 * (0.20)	0.61 (0.13)
Fractal Dimension M/L ( )	1.58 (0.04)	1.58 (0.06)	1.56 (0.04)	1.57 (0.07)	1.57 (0.05)	1.56 (0.06)	1.57 (0.05)	1.54 (0.06)
Fractal Dimension A/P ( )	1.53 (0.06)	1.49 (0.07)	1.50 (0.06)	1.49 (0.05)	1.52 (0.07)	1.51 (0.06)	1.47 * (0.08)	1.48 (0.06)

* significant (p < 0.05) within-group differences between the same conditions before and after the entire training period (in bold print for better visibility).

.

There were also main effects of time that elapsed from pre- to post-training balance assessment on the COP parameters. All significant changes occurred in the AP plane and on firm surface only and indicated a decrease in the COP measures between the first and second measurements (Table 2). The main effects of time were observed on the following sway measures: frequency (F (1.38) = 5.06, p = 0.0301), entropy (F (1.38) = 6.61, p = 0.0140), and fractality (F (1.38) = 6.11, p = 0.0180). These decreases were caused exclusively by the respective changes in the control group, and the corresponding p-values for the Tukey post hoc test were p = 0.0005 for COP frequency, p = 0.0410 for COP entropy, and p = 0.0074 for COP fractality.

4. Discussion

The purpose of this study was to determine the effects of low-intensity resistance training on a force-matching task and balance in young, healthy men. Two findings seem of particular interest. First, the applied low-intensity resistance training program improved the force accuracy expressed by AE and CE but did not affect the steadiness (variability) of the force reproductions (VE). Second, resistance training did not cause any changes in stabilographic parameters that could account for the improvement in body balance.

The results are broadly consistent with the results of our previous research on the effects of high-intensity strength training on force-matching task performance [11]. In that study, we observed that 12 weeks of high-intensity strength training led to a significant reduction in all measured force errors (AE, CE, and VE). Although the present study found that AE and CE decreased after low-intensity resistance training, there was no change in VE. AE reflects the average error from the target and measures performance accuracy, but VE serves as an indicator of the variability of the performance and provides no context for the target [21]. It can therefore be concluded that low-intensity resistance training resulted in the participants being able to reproduce the desired force value more accurately, but it did not affect the repeatability of this task.

Based on available knowledge, we are unable to identify potential causes of divergent changes in these capabilities. When analyzing changes in CE in the experimental group, the median CE for both limbs fluctuated between −40 N and −50 N before training, so the subjects were generally “under the target”. After training, the median CEs were 3 N for the right limb and −17 N for the left, respectively, which were close to zero in both cases. Lower CE values may indicate that the subjects in individual trials were closer to the target, but theoretically they could be very much over or under the target, provided that the number of over- and under-shots was similar. However, the reduction in AE after training seems to confirm the first variant. Therefore, it can be assumed that a decrease in the CE value after training also indicates an improvement in accuracy during the force-matching task.

The search for a link between resistance training and postural control yielded two important results. First, EXP did not reveal any effect of time (i.e., of training) on its results. It can therefore be argued that the applied training did not improve subjects’ stability nor affect their postural strategies. This result is not unexpected, being in accord with other authors who found neither an association between leg strength and balance [13] nor an effect of strength training on balance [22]. The latter authors used 8-week training, which resulted in a 12–17% increase in leg strength without any noticeable progress in postural control.

Second, CON showed a significant decline in postural strategies measured over time in the A/P plane on a firm surface. This means that there must have been a factor between the two tests that, acting differently on the two groups, resulted in significant differences in their final postural strategies. However, these changes do not relate to steadiness but to postural strategies, which so far have not been taken into account in similar research. This study is the first to present such results. As the first test was carried out in November and the second one in March, a possible seasonal balance variability seems a fairly likely candidate here. It has been shown, especially in men, that physical activity in the winter is the lowest [23].

Incidentally, very similar decreases in non-linear parameters and frequency of sway were caused by the transition from hard to foam surface, i.e., when the test conditions were changed from easier to more difficult by imposing additional constraints on postural control in both groups [24]. This may indicate that the lower values of non-linear parameters that CON used on a firm surface during the second test were more suited to standing on foam. Thus, it seems plausible that CON modified its strategies as if expecting greater difficulties in performing the same two tests as at baseline. It is likely that the CNS in CON became aware of the gradual formation of a sensorimotor deficit over time, which led to the observed changes in postural strategies towards their better adaptation to these deficits. However, there were no such issues with EXP, and their strategies remained unchanged. As a result, although EXP did not improve their results, as we hypothetically assumed, their systematic resistance training allowed them to avoid a temporary imbalance and to keep their postural strategies unchanged until the end of the experiment. In a similar vein, other authors have shown that frequent sensory stimulation resulting from regular physical activity leads to better functioning of balance control mechanisms [25,26,27].

It is worth noting the importance of additional information provided by non-linear measures and the COP frequency [24]. They relate directly to postural strategies that have not been investigated by earlier authors in the context of this study. Our results clearly indicate no change in postural strategies in EXP after 12 weeks of resistance training and a modification of these strategies in CON after relative physical inactivity. This observation leads to the need to modify the existing views on the relationship between physical activity and the ability to adapt postural strategies to manipulated environmental demands. The advantage of active people over others may lie in the greater flexibility of the former’s postural strategies, i.e., the ability to use them under different conditions of uncertainty [27]. On the other hand, sedentary people, who are less experienced in correctly matching strategies to tasks, try to find the right solution and, out of excessive caution, reach for the strategies needed for more demanding tasks [12]. This can lead to suboptimal choices, as evidenced by the results of this study.

Summarizing the results, it can be concluded that 12-week low-intensity resistance training improved the accuracy of the force application but did not affect the balance. We tried to create conditions for measuring balance in such a way as to provoke the greatest possible share of feedback from lower limb proprioceptors in the performance of this task (standing on thick foam). In our opinion, the lack of improvement in balance may most likely be due to two reasons, i.e., either resistance training did not improve proprioceptive feedback, or the conditions for measuring balance were not demanding enough for young trained men to manifest possible changes in balance. Assuming the truth of the first hypothesis about the lack of effect of resistance training on proprioceptive feedback, one might suspect that the observed improvement in force accuracy was due to an improved sense of effort generated centrally rather than improved proprioceptive feedback. This reasoning would be consistent with the conclusions of earlier studies that force buildup is primarily based on the efferent copy and proprioceptors play little to no role during unilateral remembered force reproduction tasks [8]. To verify the second hypothesis, it would be necessary to use more challenging balance measurement conditions in future studies.

One of the potential limitations of our work may be the easy balance conditions, which may not have been demanding enough for trained young people. Thus, the lack of improvement in postural steadiness might have been due to the ceiling effect. In future studies, it is worth making the postural task more difficult, e.g., in a single-leg stance or using sensory manipulations. Another limitation is the lack of inclusion of women in our research. Some limitations may also be due to the insufficient sensitivity of the COP parameters used to differences in postural strategies.

5. Conclusions

The 12-week low-intensity resistance training improved the accuracy of the force-matching task, as manifested by a significant decrease in absolute and constant error values. The values of the variable error characterizing the consistency of results remained unchanged. The same training did not improve postural stability or change postural strategies. However, the control group showed adverse changes in postural strategies (reduced entropy, frequency, and fractality of COP) after 12 weeks. These may indicate excessive caution in performing a simple postural task. On the other hand, the corresponding results of the EXP group after training did not differ from those obtained at the beginning of the study. This suggests that increased leg strength and/or systematic training mitigated these unfavorable effects in the study group.