Balance training (BT) can be defined as a training regimen that aims at an improved postural control (12). Balance training has been used in different age and patient groups to enhance various neuromuscular capacities (e.g., balance ability, maximal and explosive force production etc.) (15,18), for the rehabilitation of sports injuries (6,13), and in studies aiming at fall prevention (2,10). When conducting BT, it is assumed that the neuromuscular system adapts specifically and progressively to the applied training intensity (3,15,17). It has frequently been shown that a reduced base of support or deprived visual control results in a larger postural sway (1,8,14). With respect to training specificity, Silsupadol et al. (31), for example, were able to show that dual-task BT was superior to single-task training in improving walking under dual-task conditions. In fact, only participants who received dual-task BT (i.e., standing on compliant surfaces while naming objects, remembering number, etc.) walked significantly faster after the 4 weeks' training period when simultaneously performing a cognitive task. Further, comparison of a so-called ‘perturbation based’ BT with a traditional type of BT revealed significant improvements in reaction time during perturbed walking in the perturbation-based exercise group but not in the traditional exercise group (30).
To appropriately challenge the sensorimotor system during BT and to induce more profound neuromuscular adaptations, it seems advisable to alter the training intensity. This could be done by increasing the duration of exercises or the number of sets. Additionally, balance exercises could be created where the environmental conditions are gradually changed. The American College of Sports Medicine (7) suggested to reduce the base of support (e.g., from 2-legged over tandem to 1-legged stance) or to manipulate the sensory input (e.g., standing with eyes open vs. closed or on a stable vs. unstable surface).
Recent studies on the progression of exercises or activities were mainly conducted using strength-oriented tasks such as jumps, squats, or isometric contractions (4,29,32). However, to the best of our knowledge, there is at this point no scientific evidence of an optimal exercise sequence to ensure progression in BT and the impact of such a sequence on balance performance. Thus, the objective of this study was to investigate the effects of different balance exercises with a graded level of difficulty on balance performance. From an exercise-specific perspective, therapists and coaches will be provided with helpful information on the difficulty level of certain balance tasks. Thus, exercises can be chosen, which are appropriate to a subject's individual degree of expertise. It was hypothesized that the manipulation of sensory and stance conditions results in an increased level of task difficulty, which leads to enlarged postural sway.
Experimental Approach to the Problem
A single-group, repeated measure design was used to address the lack of empirical evidence about the progressive nature of balance exercises in young healthy adults using a computerized balance platform. The participants performed 3 testing sessions (independent variable) separated by approximately 24 hours. Within each session, the participants completed 4 different stance conditions, that is, bipedal, step, tandem, and monopedal stance. In session 1, stance conditions were conducted on firm ground (FI) with eyes opened (EO), on day 2, on foam ground (FO) using a balance pad with eyes opened, and on day 3, on firm ground with eyes closed (EC). Frequently used balance measures (e.g., displacement, speed, area of postural sway) were calculated to rank the different balance exercises according to their degree of difficulty from lowest to highest (i.e., low, middle, high amount of postural sway).
Twenty subjects participated in this study (9 men, 11 women; age: 22 ± 3 years; mass: 67 ± 11 kg; height: 174 ± 9 cm; body mass index: 22 ± 2 kg·m−2). All the participants, except one, were right-foot dominant as determined by the lateral preference inventory (11). None had any history of musculoskeletal, neurological, or orthopedic disorder that might have affected their ability to execute the balance exercises. The participants were physically active (∼5 h·wk−1), being involved in sports such as soccer, basketball, volleyball, or swimming. Participants' written informed consent was obtained before the start of the study. Local ethical permission was given, and all experiments were conducted according to the latest version of the declaration of Helsinki. The study was conducted from January to March 2010.
Displacements of the center of pressure (CoP) in the mediolateral and anterior-posterior directions were recorded using a computerized balance platform (HUR BT4®, HURLABS, Tampere, Finland). Time series signals were filtered using a second-order Butterworth low-pass filter with a cut-off frequency of 10 Hz. Data were acquired for 30 seconds at a sampling rate of 400 Hz. Five parameters were computed from the time series of the CoP displacements: first, the total displacements of the CoP, which represent the summed displacements in mediolateral and anterior-posterior directions (CoP_tot in millimeters); second, the displacements of the CoP in anterior-posterior direction (CoP_ap in millimeters); third, the displacements of the CoP in the mediolateral direction (CoP_ml in millimeters); fourth, the CoP speed, which indicates the total distances covered by the CoP divided by the duration of the sampled period (CoP_speed in millimeters per second); and fifth, the CoP C90area, which represents the surface area covered by the trajectory of the CoP with a 90% confidence interval (CoP_C90area in millimeters squared). The reliabilities within and between testing sessions were calculated for all balance parameters using intraclass correlation (ICC) coefficients. The ICC values ranged between 0.59 and 0.97 (26).
A single-group, repeated measure design was used to investigate the effect of balance exercises with different levels of difficulty on postural control. The participants performed 3 testing sessions, with a minimum of 24 hours in between (Figure 1). During each testing day, 4 different stance conditions were performed, that is, bipedal, step, tandem, and monopedal stance. The order of stance conditions tested was randomized between days and participants. On day 1, stand conditions were conducted on firm ground with eyes opened, on day 2, on foam ground using a balance pad (Airex®, Aalen-Ebnat, Germany) with eyes opened, and on day 3, on firm ground with eyes closed. Test circumstances (e.g., room illumination, temperature, noise) were in accordance with the recommendations for posturographic testing (21). For experimental testing, the participants were asked to stand as stable as possible on the balance platform. Before the testing started, the participants were requested to perform 2–3 practice trials to get accustomed to the testing procedure. Thereafter, three 30-second trials per stance condition were performed during each of the 3 testing days separated by a 30-second break between trials and by a 1-minute break between exercises (16,24,28). The mean of the 3 trials was used for further analysis. All balance tests were performed without shoes to allow optimal proprioceptive input and under the guidance of the same instructor. For each participant, testings were conducted at the same time of the day.
Measures of central tendency and spread of the data were represented as mean plus SD. A 1-way analysis of variance (ANOVA) with repeated measures on sensory conditions was used to detect the differences between standing on firm ground with eyes opened vs. foam ground with eyes opened vs. firm ground with eyes closed (i.e., comparison between testing days). Furthermore, a 1-way ANOVA with repeated measures on stance conditions was performed to reveal the differences between bipedal vs. step vs., tandem vs. monopedal stance within each testing day. Post hoc tests with the Bonferroni-adjusted α were conducted to identify comparisons that were statistically significant. In addition, a 1-way ANOVA across all the conditions was conducted to determine which trend component (linear, quadratic, or cubic trend line) describes the observed pattern of balance performance best (i.e., showing the highest amount of variance explained). Data are reported by the F values of the respective analysis and their p values. Moreover, η2 (eta squared) was calculated to show the amount of variance explained. Furthermore, using η2 effect sizes (f) were calculated (5) and classified as small (f values = 0.10), medium (f values = 0.25), and large (f values = 0.40) effects (9). Finally, coefficient of variation (CV) was calculated according to the following formula [(SD/Mean) × 100] (22). Using the 3 trials per testing condition, the CV represents a measure of intraparticipant trial-to-trial variability, whereas the smaller the CV value, the better the balance performance consistency. The significance level was set at α = 5%. All analyses were performed using Statistical Package for Social Sciences (SPSS) version 17.0.
The CoP displacements in the mediolateral and anterior-posterior directions for a representative participant performed during testing day 1 are given in Figure 2A–D. Comparison of sensory conditions (i.e., between testing days) revealed that total CoP displacements increased when the use of sensory information was gradually reduced, from firm ground to foam ground and from eyes opened to eyes closed (Figure 3A). The observed main effect of sensory condition was statistically significant, F(2, 38) = 139.1, p < 0.001, η2 = 0.88, f = 2.71. Post hoc analysis indicated statistically significant differences between all 3 sensory conditions (all p < 0.001). Comparison of stance conditions (i.e., within each testing day) showed that CoP displacements increased when the base of support was gradually reduced (Figure 3B). A significant main effect of stance condition was observed for day 1, F(3, 57) = 372.5, p < 0.001, η2 = 0.95, f = 4.36; day 2, F(3, 57) = 199.8, p < 0.001, η2 = 0.91, f = 3.18; and day 3, F(3, 57) = 196.2, p < 0.001, η2 = 0.91, f = 3.18. Irrespective of testing day, post hoc analysis revealed statistically significant differences between all stance conditions (all p < 0.001). The observed pattern of balance performance across all the testing conditions was best described (i.e., highest amount of variance explained) by a linear trend line with a positive slope, η2 = 0.94 (Figure 3B). This pattern of an increased postural sway when the base of support and the sensory input was gradually reduced was found irrespective of balance measure considered (Table 1). Figure 4 shows the intraparticipant trial-to-trial variability for all testing conditions. The variability ranged from 5 to 13% and was the highest for the tandem stance performed on firm ground with eyes closed and lowest for the bipedal stance performed on firm ground with eyes opened.
Using a computerized balance platform, this study quantified common balance exercises used in rehabilitative and preventive training settings. In young healthy adults, CoP displacements of 12 exercises were measured and 5 balance variables were analyzed to suggest an appropriate progression of training intensity during BT. Irrespective of balance measure, performance deteriorated because sensory input was limited (i.e., standing on a firm vs. foam ground; standing with eyes opened vs. closed) or the support surface size was reduced (i.e., bipedal, step, tandem, monopedal stance). This suggests that small changes in the base of support or the sensory input have a major effect on balance measures tested in young healthy subjects. Scientific evidence for the challenging nature of limited sensory information for balance control comes from studies using the Sensory Organization Test (SOT) (25). This test procedure is targeted on the progressive increase in manipulation of visual and kinesthetic stimuli during different conditions. Applying this protocol, Cohen et al. (8) were able to show that it is less difficult for young adults (mean age: 30 ± 8 years) to perform 2-legged standings with eyes opened on a firm ground than with eyes closed or on a foam ground. In addition, Hytonen et al. (20) reported increases in body sway velocities in young adults (age range: 16–30 years) during bipedal stance with eyes closed as compared with the eyes opened condition. Increases in sway velocity were also observed when the balance task was performed on a foam compared with when performed on a firm surface. Furthermore, Wahl and Behm (32) showed increased activity of lower extremity muscles when the standing surface was changed from stable to unstable.
The observed increase in postural sway while reducing the base of support is consistent with the findings of a study by Amiridis et al. (1). These authors investigated CoP displacements with participants (mean age: 20 ± 2 years) standing in normal quiet, Romberg-sharpened, and 1-legged stance. Amiridis et al. (1) found increases in postural sway as a result of narrowing the base of support coming along with higher levels of ankle and hip muscle activity. Combining the effect of reduced sensory information and reduced base of support on postural control, Era et al. (14) investigated normal standing with eyes open vs. closed and semitandem vs. tandem stance with eyes open in young adults (age range: 30–39 years). A change in test condition from eyes opened to eyes closed and from semitandem to tandem stand increased the speed of the CoP.
What are the underlying reasons for increased postural sway when concomitantly reducing the base of support or sensory input? In the first case, one may argue that there is a need for a longer lever arm to ensure the development of appropriate balance corrections with the aim to encounter the increased instability when, for example, changing from tandem to monopedal stance (27). Therefore, the center of mass will oscillate over the natural vertical line and as a result postural sway will increase. In the second case, it can be argued that (a) while standing with closed eyes, the visual system is useless for the control of balance resulting in a higher amount of postural sway and that (b) while standing on a foam surface instability is provided in the anterior-posterior and the mediolateral direction provoking enlarged postural sways. As a consequence, a change in balance strategy might be possible. More specifically, the successful recovery of balance demands the center of mass to remain within the boundaries of the base of support. This compensatory mechanism can be achieved by different postural movement strategies (i.e., ankle, hip, step strategy) (19). Normally, adults use the ankle strategy (i.e., distal to proximal sequence of muscle activation) to successfully reestablish their equilibrium when postural demand is low (e.g., standing on a large or firm surface) and only a slight compensatory regulation is necessary. When reducing the base of support or limiting the use of sensory information, the postural demand is increased (e.g., standing on a small or foam surface). To compensate for this higher demand in postural control, a change to hip strategy (i.e., proximal to distal sequence of muscle activation) would be appropriate because a larger degree of stabilization is provided. In this regard, Horak and Nashner (19) were able to show that subjects between the ages of 20 and 40 years change their postural movement strategy (from ankle to hip strategy) when the support surface was progressively shorten from long to short. Additionally, Kuo et al. (23) reported that healthy adults (age range: 22–35 years) increased the use of the hip strategy (significant longer sway ellipses) in the SOT conditions 5 (i.e., eyes closed, platform sway referenced) and 6 (i.e., both vision and platform sway referenced).
The increases in postural sway across testing conditions were best described by a linear trend line with a positive slope. From this, it seems possible to categorize the balance exercises used in this study by their level of difficulty. Assuming a rehabilitation program consisting of several stages, bipedal and step stance could be used in the early stage of this process, irrespective of sensory conditions considered. During the midstage of rehabilitation, the tandem and the monopedal stance performed with eyes open on a firm or foam surface could be used. Finally, the tandem and monopedal stance performed with eyes closed could be used during the late training stage. This kind of categorization is supported by the observed trial-to-trial variability in balance performance. Variability was high (i.e., warrants caution in using these exercises and conditions) especially during the tandem and monopedal stance in the eyes closed condition and was low during the bipedal and step stance. A comparison between recommendations to progressively design BT programs and the exercise sequence elicited in this study highlight the significance of our data. For example, the ACSM exercise prescription guidelines recommend challenging activities such as gradually reduced base of support or sensory input in a separated way (7), whereas the empirical evidence gathered in this study highlights the combined use of stance and sensory manipulations in the early, mid, and late stages of BT.
Guidelines concerning the optimal sequence and impact of balance exercises on postural control during BT are rare and lack of scientific validation. As an initial step, we investigated the impact of a reduced base of support and a limited sensory input on balance performance. The displacements increased when manipulating the base of support (i.e., bipedal, step, tandem, monopedal stance) and the sensory information (i.e., eyes opened vs. closed; firm vs. foam ground) available. Depending on the size and variability of postural sway, exercises could be categorized in easy (e.g., bipedal, step stance irrespective of sensory condition), mid (e.g., tandem, monopedal stance with eyes opened on a firm and foam ground), and hard (e.g., tandem, monopedal stance with eyes closed) to perform for young healthy adults. Findings can be used by practitioners to select exercises for BT appropriate to the level of participants' balance ability.
1. Amiridis, IG, Hatzitaki, V, and Arabatzi, F. Age-induced modifications of static postural control
in humans. Neurosci Lett
350: 137–140, 2003.
2. Barnett, A, Smith, B, Lord, SR, Williams, M, and Baumand, A. Community-based group exercise improves balance and reduces falls in at-risk older people: A randomised controlled trial. Age Ageing
32: 407–414, 2003.
3. Beck, S, Taube, W, Gruber, M, Amtage, F, Gollhofer, A, and Schubert, M. Task-specific changes in motor evoked potentials of lower limb muscles after different training interventions. Brain Res
1179: 51–60, 2007.
4. Behm, DG, Anderson, K, and Curnew, RS. Muscle force and activation under stable and unstable conditions. J Strength Cond Res
16: 416–422, 2002.
5. Bortz, J. Statistik für Sozialwissenschaftler
. Berlin, Germany: Springer Verlag, 1999.
6. Caraffa, A, Cerulli, G, Projetti, M, Aisa, G, and Rizzo, A. Prevention of anterior cruciate ligament injuries in soccer. A prospective controlled study of proprioceptive training. Knee Surg Sports Traumatol Arthrosc
4: 19–21, 1996.
7. Chodzko-Zajko, WJ, Proctor, DN, Fiatarone Singh, MA, Minson, CT, Nigg, CR, Salem, GJ, and Skinner, JS. American college of sports medicine position stand. Exercise and physical activity for older adults. Med Sci Sports Exerc
41: 1510–1530, 2009.
8. Cohen, H, Heaton, LG, Congdon, SL, and Jenkins, HA. Changes in sensory organization test scores with age. Age Ageing
25: 39–44, 1996.
9. Cohen, J. Statistical Power for the Behavioral Sciences
. Hillsdale, NJ: Erlbaum, 1988.
10. Conroy, S, Kendrick, D, Harwood, R, Gladman, J, Coupland, C, Sach, T, Drummond, A, Youde, J, Edmans, J, and Masud, T. A multicentre randomised controlled trial of day hospital-based falls prevention programme for a screened population of community-dwelling older people at high risk of falls. Age Ageing
39: 704–710, 2010.
11. Coren, J. The lateral preference inventory for measurement of handedness, footedness, eyedness, and earedness: Norms for young adults. Bull Psychon Soc
31: 1–3, 1993.
12. DiStefano, LJ, Clark, MA, and Padua, DA. Evidence supporting balance training in healthy individuals: A systemic review. J Strength Cond Res
23: 2718–2731, 2009.
13. Emery, CA, Rose, MS, McAllister, JR, and Meeuwisse, WH. A prevention strategy to reduce the incidence of injury in high school basketball: A cluster randomized controlled trial. Clin J Sport Med
17: 17–24, 2007.
14. Era, P, Sainio, P, Koskinen, S, Haavisto, P, Vaara, M, and Aromaa, A. Postural balance in a random sample of 7,979 subjects aged 30 years and over. Gerontology
52: 204–213, 2006.
15. Granacher, U, Gollhofer, A, and Kriemler, S. Effects of balance training on postural sway, leg extensor strength and jumping height in adolescents. Res Q Exerc Sport
81: 245–251, 2010.
16. Granacher, U, Muehlbauer, T, Taube, W, Gollhofer, A, and Gruber, M. Sensorimotor training. In: Strength and Conditioning: Biological Principles and Practical Applications
. M. Cardinale, R. Newton, and K. Nosaka, eds. Chichester, United Kingdom: John Wiley & Sons Limited, 2011. pp. 399–409.
17. Gruber, M, Bruhn, S, and Gollhofer, A. Specific adaptations of neuromuscular control and knee joint stiffness following sensorimotor training. Int J Sports Med
27: 636–641, 2006.
18. Gruber, M and Gollhofer, A. Impact of sensorimotor training on the rate of force development and neural activation. Eur J Appl Physiol
92: 98–105, 2004.
19. Horak, FB and Nashner, LM. Central programming of postural movements: Adaptation to altered support-surface configurations. J Neurophysiol
55: 1369–1381, 1986.
20. Hytonen, M, Pyykko, I, Aalto, H, and Starck, J. Postural control
and age. Acta Otolaryngol
113: 119–122, 1993.
21. Kapteyn, TS, Bles, W, Njiokiktjien, CJ, Kodde, L, Massen, CH, and Mol, JM. Standardization in platform stabilometry being a part of posturography. Agressologie
24: 321–326, 1983.
22. Kressig, RW and Beauchet, O. Guidelines for clinical applications of spatio-temporal gait analysis in older adults. Aging Clin Exp Res
18: 174–176, 2006.
23. Kuo, AD, Speers, RA, Peterka, RJ, and Horak, FB. Effect of altered sensory conditions on multivariate descriptors of human postural sway. Exp Brain Res
122: 185–195, 1998.
24. Le Clair, K and Riach, C. Postural stability measures: What to measure and for how long. Clin Biomech
11: 176–178, 1996.
25. McCollum, G, Shupert, CL, and Nashner, LM. Organizing sensory information for postural control
in altered sensory environments. J Theor Biol
180: 257–270, 1996.
26. Muehlbauer, T, Roth, R, Mueller, S, and Granacher, U. Intra and intersession reliability of balance measures during 1-leg standing in young adults. J Strength Cond Res
27. Nejc, S, Jernej, R, Loefler, S, and Kern, H. Sensitivity of body sway parameters during quiet standing to manipulation of support surface size. J Sports Sci Med
9: 431–438, 2010.
28. Pinsault, N and Vuillerme, N. Test-retest reliability of centre of foot pressure measures to assess postural control
during unperturbed stance. Med Eng Phys
31: 276–286, 2009.
29. Scarfe, AC, Li, FX, Reddin, DB, and Bridge, MW. A new progression scale for common lower-limb rehabilitation tasks. J Strength Cond Res
25: 612–619, 2011.
30. Shimada, H, Obuchi, S, Furuna, T, and Suzuki, T. New intervention program for preventing falls among frail elderly people: The effects of perturbed walking exercise using a bilateral separated treadmill. Am J Phys Med Rehabil
83: 493–499, 2004.
31. Silsupadol, P, Shumway-Cook, A, Lugade, V, van Donkelaar, P, Chou, LS, Mayr, U, and Woollacott, MH, et al. Effects of a single-task versus dual-task training on balance performance in older adults: A double-blind, randomized controlled trial. Arch Phys Med Rehabil
29: 634–639, 2009.
32. Wahl, MJ and Behm, DG. Not all instability training devices enhance muscle activation in highly resistance-trained individuals. J Strength Cond Res
22: 1360–1370, 2008.
Keywords:Copyright © 2012 by the National Strength & Conditioning Association.
postural control; task difficulty; base of support; sensory input