The Effect of an Eyes-closed Dance-specific Training Program on Dynamic Balance in Elite Pre-professional Ballet Dancers A Randomized Controlled Pilot Study Kimberley Hutt, M.Sc., and Emma Redding, Ph.D. Abstract
Visual conditions for a dancer vary greatly between theatrical performance environments and dance studios, and this variability may be detrimental to their dynamic balance performance, particularly under stage lighting. In order to maintain balance control, dancers reportedly rely heavily on visual input, yet those who rely more on proprioceptive strategies for balancing have been found to be more stable. The purpose of this study was to assess the capability of an eyes-closed, dancespecific training program to nurture in dancers proprioceptive mechanisms that may facilitate their dynamic balance control. Eighteen elite pre-professional ballet dancers were randomly assigned to either a control (eyes open) or experimental (eyes closed) group for the intervention. The balance abilities of all subjects were tested using five dance-specific variations of the Star Excursion Balance Test before and after a 4 week balance intervention. Reach distance and time to complete the tests were recorded separately as indirect measurements of dynamic balance. The intervention consisted of dance-specific, eyes-closed exercises integrated into the dancers’ daily ballet class and designed progressively to challenge the dancers’ balance. During the intervention period, the control group undertook the same exercise program with their eyes open. Results revealed significant improvements in time to complete the three “timed” balance
tests, and distances reached significantly improved in one of the two “reach” balance tests. No significant improvements were observed in the control group for any variation of the tests. These results indicate that dancers can be trained to adopt proprioceptive strategies to maintain dynamic balance, which consequently improves their balance performance. Such findings could encourage use of eyes-closed training in daily dance classes due to its potential to improve dancers’ balance control.
C
lassical ballet has attracted audiences worldwide since the 17th Century.1 It is an art form that becomes more accessible when its performers have mastered the fundamental skills required to execute complex movement patterns. One such skill is balance. Balance is not only essential for successful ballet performance,2 but many classical choreographies showcase a dancer’s balance ability to excite an audience—for example, Princess Aurora in Sleeping Beauty. Classical ballet dancers must have sophisticated balance mechanisms in order to position themselves effectively during the complex choreographed sequences of dance practice and performance. Dance involves multidi-
Kimberley Hutt, M.Sc., and Emma Redding, Ph.D., are at the Trinity Laban Conservatoire of Music and Dance, London, United Kingdom. Correspondence: Kimberley Hutt, M.Sc., 42 Laurel Avenue, Potters Bar, Hertfordshire EN6 2AB, United Kingdom; [email protected]. Copyright © 2014 J. Michael Ryan Publishing, Inc. http://dx.doi.org/10.12678/1089-313X.18.1.3
rectional and rotational activities that challenge balance control, and dancers continually strive to improve balance in order to enhance their stability during the execution of a step or piece of choreography. Yim-Chiplis and Talbot3 stated that balance provides the foundation for human mobility and functional independence, highlighting the fundamental importance of balance to normal human locomotion. Research has shown that dance practice itself can autonomously provide effective balance training. 4 While the precise neurophysiological mechanism behind this effect is yet to be fully established, dancers have been shown to be exceptionally sensitive to variations in their gravitational axis.5 Consequently, research has used dance as a method to improve the balance ability of non-dancers6; however, dancers’ balance abilities need to be superior to those of the general population for successful and efficient dance performance. Balance refers to the ability to maintain the body’s center of gravity over its base of support.7 During dance, when the center of gravity is displaced beyond the vertical projection of the base of support, the dancer will become unstable and begin to fall.8 This explains why dynamic balance becomes more difficult as the base of support is reduced in size, for example, from fifth position to arabesque, from a flat foot to demi 3
4
Volume 18, Number 1, 2014 • Journal of Dance Medicine & Science
pointe, or en pointe as seen in classical ballet. To maintain balance, there must be an efficient and complex integration of the visual, vestibular, and somatosensory organs.9 Within these organs, there are specialized sensory receptors which provide information about the external environment (exteroceptors), maintain physiological homeostasis (interoceptors), and report on the position and movement of muscles and joints (proprioceptors).10 Proprioceptors within muscles and tendons are not only responsible for detecting body movement and position but also contribute to the control of postural reflexes to maintain stability while moving11 and are paramount in the prevention of injury due to their role in joint and postural stability. It seems prudent, therefore, to increase the challenge to these specialized receptors during dance practice with the aim of enhancing dancers’ proprioception and consequent balance abilities and potentially to reduce the risk of injury associated with loss of stability. There is general consensus in the literature that ballet dancers use proprioceptive and visual stimuli as their fundamental sensory inputs to maintain balance. Although as yet only anecdotal, research has suggested that due to the habitual use of mirrors during ballet class and rehearsals dancers rely too heavily on visual cues for dynamic balance while dancing.12 This could become problematic to a dancer’s balance when those visual cues change under different dance conditions. Based on reports of dancers’ dependence on visual input for balance control and when considering that visual conditions in theatrical performance settings are not representative of those in dance studios, the unfamiliar and unaccommodating visual conditions encountered in a theater may prove detrimental to dynamic balance during a dancer’s performance.2 Research has reported an ability for ballet dancers to modify the sensory organization that is responsible for balance control.13,14 This modification is known as a “shift,” and it refers to the transition from using one balance
mechanism to another. For example, dancers may use visual input as the dominant mechanism to maintain balance when their eyes are open, yet acutely shift to more proprioceptive strategies when their eyes are closed. A similar shift might be required under stage lighting conditions. As previously noted, classical ballet dancers’ balance abilities are reportedly superior to non-dancers’, yet when their eyes are closed, their balance abilities are no better,2 suggesting that their ability to shift acutely from one balance mechanism to another is unsophisticated. This could be explained by Hugel’s hypothesis that the practice of dance alone does not enhance the brain’s afferent signals of vestibular and proprioceptive stimulus.2 This in turn may be explained by the heavy reliance on visual stimulus and stable environments seen in dance practice. Perrin and coworkers15 found that Judokas have superior balance abilities compared to classical ballet dancers when tested with their eyes closed. The investigators inferred that Judo relies more heavily on proprioception than visual references, which may be advantageous to balance control. Therefore, research is needed to investigate the effects of balance training that favors proprioception as the dominant balance mechanism, with the aim of producing a chronic shift in sensory organization from an external visual dependency to a more proprioceptive “internal awareness” of stability when dancing. In their consideration of balance training for ballet dancers, Huxham, Goldie, and Patla16 stated that balance “cannot be separated from the action of which it is an integral component, or from the environment in which it is performed.” Willardson17 supported this theory, stating that balance is skill specific, and that it should be trained using the same technical skills that are required during performance. Therefore, in training dancers it seems appropriate to select a balance-training program with specificity to dance practice. The principle of specificity in sport is well documented and acknowl-
edged. 18 Considering the unique nature of dance and the purported benefits of specificity, it is no wonder that in the growing field of dance science researchers are developing dance-specific testing protocols and training regimens for a number of physiological, biomechanical, and aesthetic components of dance.19,20 Balance and stability testing must also be specific to the skills of the performer. The Star Excursion Balance Test (SEBT) has been said to stress dynamic balance control21-23 and is being used among the athletic population.24 The SEBT is arguably, therefore, more appropriate for testing dancers’ balance than other validated static balance tests. On the other hand, while the SEBT involves reaching a leg in multiple directions that replicate those seen in dance practice (for example, croisé and ouvert), it arguably lacks skill specificity to dance practice. In fact, while balance tests such as the SEBT are routinely implemented in screening and rehabilitation programs among the athletic and dance population, until recently no balance test has been designed with specificity to a given sport or activity, such as dance. The SEBT has now been modified (mSEBT) to improve its specificity to dance practice,25 and since the creation and initial testing of the mSEBT, researchers Glenna Batson and Margaret Wilson have been working collaboratively to improve the test.26 Developments to date have included randomization of the direction to which the working leg extends, in order to challenge the attentional focus of the dancer. As with the mSEBT, the “randomized” Star Excursion Balance Test (rSEBT) records measurements of reach distance and timings. While the randomized modifications have yet to be validated and published, the present investigators consider these further modifications valuable for their specificity to dance practice, due to the rapid responses to directional cues required when performing complicated dance sequences. The objective of the present study was to improve the dynamic balance
Journal of Dance Medicine & Science • Volume 18, Number 1, 2014
performance of a sample of dancers by increasing the challenge to their proprioceptive mechanisms of balance control in their daily dance practice. It was anticipated that the increased challenge could be achieved by integrating an eyes-closed training program to “switch off” the visual mechanisms, thus instigating a chronic shift from visual reliance to proprioceptive strategies for the maintainance of dynamic balance. The researchers hypothesized that incorporating eyesclosed practice into daily ballet classes would enhance compliance with the program and reduce the necessity for additional training. Significant improvements in dymanic balance abilites would provide sound rationale for dance educators to promote eyes-closed training in dancers’ daily practice.
Materials and Methods
Subjects The selection of subjects and protocol of the study were ethically approved prior to data collection and the intervention. All hard and electronic copies of subjects’ data were stored in compliance with the Data Protection Act of 1998, and anonymity was maintained throughout the study. Nineteen elite female pre-professional dancers from Central School of Ballet (UK) volunteered and consented to participate. The sample comprised the first year group, who had all been studying full time at the school for 9 months. Prior to testing, the researcher documented the subjects’ heights using a Bodymeter Measuring Tape (Factory Direct, USA) and recorded their mass in Newtons, taken from a portable force plate (Kistler®, Switzerland, threshold Fz < 10 mN, model 9286AA) and converted to kilograms in order to calculate BMI. For the participants’ epidemiology see
Table 1. Subjects were randomly assigned to one of two groups, control or experimental, by blindly selecting a piece of paper marked with “C” (control) or “E” (experimental) from a “hat.” This resulted in 10 experimental participants and 9 controls. At the beginning of the study, all subjects met the inclusion criteria, which stated that they should have no vestibular disorder or recent history of lower limb injury. During the intervention period, one dancer in the control group sustained an injury; therefore, data were collected for 10 dancers in the experimental group and 8 in the control group. Instrumentation A star-shaped grid, seen in Figure 1, was constructed using Cramer Tape Heavy Zinc Oxide (2.5 cm wide). The portable force plate was mounted longitudinally in the center of the grid. This was connected to a Kistler® DAQ-system which transmitted information of forces to a laptop computer equipped with Bioware System software version 4.1. Methodology The researcher determined leg dominance by asking the subjects to declare their favored “working leg” when dancing. Working leg was described to participants as being that which they preferred to “practice exercises with,” as opposed to “stand on.” Leg length data were collected using measurements from the greater trochanter to the distal lateral malleoli, in order that measurements of reach could be normalized. Testing procedures took place 3 days before and 3 days after a 4 week intervention. Procedures were explained to each subject, following which five variations of the original
5
Figure 1 The eight spoke grid and directions used for all variations of the SEBT. NB: These directions are based on the subject facing the front spoke and using her right leg as the dominant “reaching” leg.
SEBT were performed by each participant using their non-dominant leg as the supporting leg and their dominant leg as the reaching leg. The same researcher undertook all explanations and data collection to enhance intra-test reliability. The Basic SEBT Subjects were instructed to stand on the force plate with their supporting leg in parallel position and their dominant leg held in a parallel coup de pied (the “ready position,”25 Fig. 2). They were then instructed to bend the supporting leg and reach toward each spoke, first in a clockwise, then counter-clockwise direction, lightly “dabbing” the toe at the furthest point they could reach. Distance reached, missed spoke, falls, and near falls were recorded on a testing sheet. The entire test was terminated if a participant hopped or fell due to loss of balance or put excessive weight on the working foot when it was “dabbed” at each spoke. Subsequent Variations of the Basic SEBT On completion of the basic SEBT, the dancers were given instructions
Table 1 Participant Characteristics at the Start of the Study Mean Height (± SD) (cm)
Mean Mass (± SD) (kg)
Mean BMI (± SD)
Mean Leg Length (± SD) (cm)
Mean Age (± SD) (yrs)
Mean Years Dancing (± SD) (yrs)
163.82
50.75
18.95
82.26
16.68
11.16
(5.55)
(4.5)
(1.2)
(4.75)
(0.48)
(3.11)
6
Volume 18, Number 1, 2014 • Journal of Dance Medicine & Science
Figure 2 The “ready” position.
on the variations of four subsequent tests and performed the following modifications: 1. Modified SEBT—timed test (mSEBTtimed): The basic SEBT was performed as quickly and safely as possible. Participants were instructed to reach as far as possible in the fastest time they could. The time taken to finish one full circuit of the test in a clockwise and counter-clockwise direction was recorded, along with missed spokes, near falls, and falls. 2. Modified SEBT—timed test with cognitive interference (mSEBTcogint): The mSEBTtimed was performed while the researcher asked a series of questions for cognitive interference; for example, “Spell your surname backwards” and “What is one hundred minus thirteen?” Dancers with English as a second language were invited to answer the questions in their own language. Again, time taken to finish one full circuit of the test in a clockwise and counterclockwise direction was recorded, along with missed spokes, near falls, and falls. 3. Randomized SEBT—reach distance test (rSEBTreach): To stan-
dardize the test between subjects, an identical “randomized” order of the direction toward which the subject should reach was used by the researchers. Subjects were instructed to reach as far as they safely could, at their own pace. The subjects were taught the names given for the directions for each spoke (Fig. 1) and told that the direction to which they should reach would be given as soon as the working foot came back to the “ready position.” Wilson recommended the names of the spokes in research using the rSEBT (unpublished). It was considered that these terms would be easier for the dancer to interpret than the scientific terms used in standard research (e.g., anterior, posterior, medial, and lateral). Distance reached, missed spoke, falls, and near falls were recorded. 4. Randomized SEBT—timed test (rSEBTtimed): The rSEBTreach was repeated as fast and as safely as possible. For this variation of the test subjects were instructed to extend their leg to approximately 45°, rather than dab their toe to the floor. The researcher recorded time taken to complete the sequence, along with missed spokes, near falls, and falls. Due to suggestions made by Batson25 for future studies to observe quantitative force plate analysis, measurements of sway around the center of pressure (COP) in anteriorposterior (AP) and medial-lateral (ML) direction were recorded during all variations of the test. The present study did not obtain data using the foam modification of the mSEBT used by Batson,25 as the force plate would not transmit useful data of forces through foam. The Intervention Prior to the study, the lead researcher met with the participants’ regular ballet teacher to discuss the balance exercises that were to be integrated into the dancers’ daily class, forming the basis of the intervention. The researcher explained that the interven-
tion should be the first center practice exercises following barre work for the sake of consistency and should be performed 5 days per week for 4 weeks. The sequences were to become progressively challenging on a weekly basis, without posing any unnecessary risk to the dancers. The researcher suggested examples of such progressions, including stationary sequences progressing to travelling and turning sequences; double leg stance exercises progressing to single leg stance exercises; and sequences involving working legs a terre progressing to sequences with the working leg en l’air. When the dance-specific exercises had been choreographed by the dance teacher and approved by the researcher, the experimental group was instructed to perform the exercises with their eyes closed throughout the intervention period. They should only open their eyes during the eyes-closed sequence if they felt they were going to fall, and in that event, they should close their eyes and continue as soon as they felt they had regained their balance sufficiently to continue safely with the sequence. Throughout the intervention period the control group undertook the same program with their eyes open, as they would under normal dance class conditions. Data Analyses Mean (± SD) measurements were calculated for distance reached (cm), time to completion (s), or sway (mm) around the center of pressure (COP), depending on the balance test being performed. Using SPSS software (version 19; IBM, USA), with alpha set to 0.05, paired samples t-tests were carried out to assess for balance improvements following the intervention. Due to the use of multiple t-tests, Benjamini Hochberg corrections were undertaken as a valid means of reducing the likelihood of creating a type 1 statistical error, following which all p values were divided by two (alpha level = 0.025) to reflect the directional nature of the hypotheses—that reach distance and time would improve. The statistical analyses used mean data taken from each of the following
Journal of Dance Medicine & Science • Volume 18, Number 1, 2014
dependent variables: 1. Basic SEBT (reach distance measured in cm), 2. mSEBTtimed (time measured in seconds), 3. mSEBTcogint (time measured in seconds), 4. rSEBTreach (reach distance measured in cm), 5. rSEBTtimed (time measured in seconds), 6. Center of pressure – anteriorposterior (sway measured in mm), and 7. Center of pressure – mediallateral (sway measured in mm). Center of pressure (COP) measurements were only analyzed from data collected during the rSEBTtimed, as this was the only variation in which the working foot did not touch the floor. The act of “dabbing” the floor during all other variations of the test rendered force plate data unusable, due to extreme inter-subject variability of pressure with which the foot was “dabbed.”
Results Both groups appeared to improve in distance reached and time to completion of the balance tests; however, the improvements were greater among the experimental participants. Statistical analysis using paired samples t-tests revealed significant improvements in the balance of the experimental group
when undertaking four of the five variations of the test: basic SEBT (t [9] = -3.256, p = 0.01), mSEBTtimed (t [9] = 3.086, p = 0.01), mSEBTcogint (t = [9] 3.458, p = 0.01), and rSEBTtimed (t [9] = 4.643, p = 0.00). Conversely, no statistically significant differences were observed in the control group when undertaking any of the variations of the balance tests (Table 2). The control group demonstrated a 0.1% deterioration (increase) in time to complete the mSEBTtimed, an 8% improvement (decrease) in time to complete the mSEBTcogint, and a 4% improvement in time to complete the rSEBTtimed following the intervention period. More impressively, the experimental group demonstrated a 14% improvement in time to complete the mSEBTtimed, a 15% improvement in time to complete the mSEBTcogint, and a 16% improvement in time to complete the rSEBTtimed test (Fig. 3). When observing reach distances achieved, the control group managed to reach an average of 4% further during the basic SEBT and 0.3% further during the rSEBTreach after the 4-week period. These results proved to be non-significant. In the experimental group, the intervention elicited a significant increase of 5% reach distance achieved during the basic SEBT, but only a 0.9% increase during the rSEBTreach, which proved
7
to be non-significant (Table 2, Fig. 4). Neither trend nor significant difference was observed in any of the center of pressure measurements tested (control group: anterior-posterior – t [5] = -1.478, p = 0.34; medial-lateral – t [5] = 1.172, p = 0.19. Experimental group: anterior-posterior – t [7] = 1.537, p = 0.12; medial-lateral – t [7] = -0.825, p = 0.22). Therefore, it cannot be said that center of pressure measurements differed following an intervention to improve balance in dancers. In summary, results suggest that due to the observed trends in and largely significant improvements in the experimental group that were not present in the control group, the dynamic balance control of elite pre-professional ballet students did improve following a period of eyesclosed dance-specific training.
Discussion The purpose of the present study was to determine whether a 4-week, eyesclosed, dance-specific training program would improve dynamic balance in a sample of elite pre-professional dancers. To date, no other published research has reported the effectiveness of an eyes-closed training program on balance abilities in dancers. The dominant statistical findings revealed that when testing the dynamic balance ability of dancers performing
Table 2 Timings and Reach Distance for All Variations of the SEBT Basic SEBT (cm)
mSEBTtimed (s)
mSEBTcogint (s)
rSEBTreach (cm)
rSEBTtimed (s)
Pre
Post
Pre
Post
Pre
Post
Pre
Post
Pre
Post
Control
Mean (± SD)
69.9 (± 3.5)
72.4 (± 4)
16.5 (± 3.1)
16.5 (± 3)
18.1 (± 4.3)
16.6 (± 3.6)
74.3 (± 5.6)
74.5 (± 3.5)
29.2 (± 3.4)
28.1 (± 2.8)
Experimental Mean (± SD)
79.4 (± 5.6)
83.4 (± 7)
17.9 (± 3.4)
15.5 (± 3)
18.1 (± 3.3)
15.5 (± 3.1)
82.6 (± 6.6)
83.4 (± 6.5)
29.2 (± 3.1)
24.7 (± 2.3)
Descriptives
Paired samples t-test Control *Significance T value
NS (p = 0.07) -3.049
NS (p = 0.49) -.038
NS (p = 0.28) 1.251
NS (p = 0.51) -.161
NS (p = 0.24) 1.182
Experimental *Significance T value
SIG (p = 0.01) -3.256
SIG (p = 0.01) 3.086
SIG (p = 0.01) 3.458
NS (p = 0.22) 0.857
SIG (p = 0.00) 4.643
*NS = no significance; SIG = significant
8
Volume 18, Number 1, 2014 • Journal of Dance Medicine & Science
Figure 3 Mean times for both groups’ timed tests.
Figure 4 Mean reach distance for both groups’ reach tests.
variations of the mSEBT and rSEBT, time to complete all variations of the balance tests significantly improved following the intervention. Reach distances significantly improved when performing the basic SEBT but not during the rSEBTreach. There were no significant improvements observed among the control group across all variations of the balance tests. Theorizing that the test variations collectively represent dynamic balance, overall these results suggest that a 4-week, eyes-closed, dance-specific training program can improve the dynamic balance ability of dance students. Reach distance and timing may not be considered direct measures of balance. However, tasks such as these create perturbations that increase the challenge to the central nervous
system to generate appropriate motor responses. Indeed, Huxham, Goldie, and Patla 16 stated that increasing the speed of a moving limb requires greater equilibrium control to maintain balance control, thus justifying the inclusion of a test of speed in the mSEBT, rSEBT, and the present study. Interestingly, all variations of the timed tests performed in the present study revealed significant improvements; therefore, it may have been the dancers’ equilibrium control component of balance that improved to the greatest extent. The reach component of the basic SEBT and rSEBTreach involves changes in the magnitude, direction, and combination of forces acting on and produced by the body. As the direction of the reach was constantly
changing, the biomechanical challenges to maintain balance control were increased. The current study found inconclusive results when considering reach as an isolated measure of balance, with only the basic SEBT variation of the reach tests (and not the rSEBTreach) revealing significant improvements. However, the additional challenge produced by randomization involved in the rSEBTreach may have caused the more variable results. These additional challenges could be explained by Huxham, Goldie, and Patla’s16 description of two mechanisms, proactive and reactive, by which environmental factors influence the biomechanical properties of balance. Proactive balance control involves information received through the eyes but also includes a form of
Journal of Dance Medicine & Science • Volume 18, Number 1, 2014
control known as predictive control. Predictive balance control relies on anticipatory postural adjustments in response to internal representations of the body and learned movement.16 In the case of predictive control, muscle activity begins before the actual movement takes place, with the magnitude of these muscle responses being dependent upon the direction and speed of the movement. The randomization of direction used in the rSEBTreach variation would have certainly disrupted anticipatory mechanisms, as the anticipatory muscle activity could not have been able to respond in anticipation of a predicted direction, a factor that could account for the non-significant improvements when performing the rSEBTreach. Despite the lack of significance, it is noteworthy that the experimental group achieved greater improvements in reach distances for the rSEBTreach than the control group, and it may only have been a matter of time before significance was achieved. Testing durations of balance studies among healthy adolescents have ranged from 4 to 7 weeks,27-30 suggesting that 4 weeks of training, as used in the present study, is sufficient time to stimulate an effect. Many of these studies reported non-compliance to the training program as a limitation of the intervention. Dancers’ daily training loads are high, and interventions that impose exercises in addition to daily practice may not be adhered to. The present study boasts excellent compliance, as the exercises were incorporated into the dancers’ morning ballet class 5 days per week for the duration of the study. Compliance is likely to be greater when the training can be incorporated into the daily training activities, as ancillary practice will rely heavily on individual dedication to the research program. Batson25 reported variable execution of the balance tests during her preliminary study; however, her study’s protocol dictated that testers should not demonstrate the test to the subjects, but rather the testers instructed the subjects by reciting from a learned script. The researchers of
the present study saw this as a limitation to Batson’s study that may have resulted in the observed inter-subject variability and consequently chose to demonstrate the balance tests to the subjects prior to data collection. Perhaps as a result of this, the execution of the tests was generally consistent among dancers. Interestingly, the greatest variability in test execution was observed among dancers who had 5 years or less dance experience. Wilmerding and Krasnow31 state that perception (the dancer’s observation and attention to a given demonstration) and execution (the dancer’s replication of the given demonstration) are essential stages involved in motor learning. These stages are likely to be less refined in dancers with fewer years of experience. Such concepts are supported by the variability of test execution seen by Batson,25 who studied dancers with a mean of only 2.6 (± 2.7) years training in participants from the UK and 9.6 (± 4.6) years training in those from the USA. However, further research would be required to confirm a relationship between variability of execution of balance tests and years of dance experience. Following suggestions for future research by Batson,25 the present study analyzed force plate data to investigate postural sway as represented by COP measurements. No trends were observed in this way, and inter- and intra-participant results elicited considerable variability. One explanation for this may be “exploratory activity” in response to a new movement pattern. However, such activity is not necessarily destablizing 32; indeed, the notion of increased sway being a reflection of instability is a common misconception, and postural sway may not be a direct indicator of postural stability.33 Alternatively, large sway may be due to deliberate actions taken to maintain balance, rather than a measure of instability.32 van Emmerick and van Wegen34 suggest that increased oscillations identified by force plate data may be an indication of greater dynamic neuromuscular function to reduce instability. The
9
investigators deduce that increased postural dynamics enhance the ability of the body to adapt to imposed postural disturbances. While the precise neurophysiological adaptations behind the balance improvements observed in the present study have not been identified, it could be postulated that they were due to a trained (chronic) shift in sensory organization from visual to proprioceptive dependency to maintain balance control. Future studies might observe EMG activity to analyze muscular activity in response to the potential shift in sensory organization.
Study Limitations The present study did not assess the range of ankle dorsiflexion available to each participant, as did Batson.25 It is conceivable that several dancers reached the limit of dorsiflexion on their supporting leg before reach distance became sufficient to disturb their balance. In this case, it is unlikely that reach measurements would change following an intervention, as it would not be balance itself that was limiting the reach distance achievable but rather the range of motion available at the ankle. The present study determined leg dominance in an unorthodox manner, by asking participants what their favored “working leg” was when practicing dance. Traditional methods of determining leg dominance are varied, and there are disparities between “functional” (for example, the kicking leg in football) and “strength” (for example, the standing leg during postural control tasks) definitions of leg dominance.33 Failure to use an established method for determining leg dominance in the present study may have resulted in variability of test results among the participants. The entire sample of dancers used in the present study was female. When testing one gender it cannot be assumed that results are transferrable to the other, yet there is no literature to suggest that balance training affects genders differently. When testing dynamic balance Golomer35 observed that males are more dependent on vi-
10
Volume 18, Number 1, 2014 • Journal of Dance Medicine & Science
sion than females; therefore, research into the effects of eyes-closed training for dancers of both genders is merited. While the present study provides valuable information about balance training in female dancers in their late teens, both the dance and scientific literature generally contain considerable discussion of the variability in balance mechanisms employed at different ages. Therefore, the results from the present study cannot be generalized to dancers of all age groups, and future research might attempt to determine whether there is an optimal age for balance training. Furthermore, mechanisms of balance control between dancers with varied years of training should be investigated. While all dancers in the present study were considered to be “elite pre-professionals” and demonstrated similar abilities in dance classes, variability of execution of the balance tests was only observed in the dancers with fewer years of dance experience. All balance tests were performed in parallel position of the supporting leg, as prescribed by the original SEBT and subsequent modifications. The dancers in the present study were all aspiring ballet dancers; as such they favored a “turned out” position and felt uncomfortable in the required parallel position. In fact, for some ballet dancers, standing with the feet in parallel requires them to rotate their femurs internally. This is due to dancers commonly demonstrating greater tibial version angles than the general population36 and increased foot progression angles (or “out toeing”).37 To maintain specificity to dance, future studies could consider allowing ballettrained subjects to stand in turned out position, more closely to replicate balance abilities in their chosen dance style. Future researchers should take note that the data collected from the present study cannot be compared to that collected without the use of a force platform. The small elevated step produced by the force platform no doubt affected the distance the dancers were able to reach.
It is noteworthy that by chance there was a difference in balance abilities between the experimental and control groups pre-intervention, despite the random selection of participants. This difference proved to be significant when analyzed statistically. The researchers could not have foreseen this anomaly; however, the results should be interpreted carefully, as the control group may well have had more sophisticated balance strategies pretesting, which may have reduced any potential effects of the intervention exercises. Nonetheless, the exercises used for the intervention were not considered to be sufficiently challenging when performed with the eyes open (the control group condition) to elicit an improvement in balance in elite level dancers.
Conclusion To the best of the researchers’ knowledge, this study is the first to assess the effectiveness of eyes-closed training on the dynamic balance abilities of dancers. The research used speed and reach distance (of a lower limb) as indirect measurements of dynamic balance, and the resulting data indicate that eyes-closed training can improve the balance abilities of elite female pre-professional ballet dancers. These results imply that closing the eyes during dance training is an effective way to stimulate a shift from visual to proprioceptive dependency for balance control, thereby improving balance regardless of visual conditions in the surrounding environment. However, further research is needed to investigate the precise neural and biomechanical mechanisms by which the improvements observed in this study were achieved. These findings may then encourage dance educators and practitioners to explore the benefits to be derived from incorporating eyesclosed training into dance practice.
References 1. Kassing G. History of Dance: An Interactive Arts Approach. Champaign, IL: Human Kinetics Publishers, Inc., 2007. 2. Hugel F, Cadopi M, Kohler F, Perrin
P. Postural control of ballet dancers: a specific use of visual input for artistic purposes. Int J Sports Med. 1999 Feb;20(2):86-92. 3. Yim-Chiplis PK, Talbot LA. Defining and measuring balance in adults. Biol Res Nurs. 2000 Apr;1(4):32131. 4. Brown S, Martinez MJ, Parsons LM. The neural basis of human dance. Cereb Cortex. 2006 Aug;16(8):115767. 5. Mouchnino L, Ayrenty R, Massion J, Pedotti A. Coordination between equilibrium and head-trunk orientation during leg movement: a new strategy built up by training. J Neurophysiol. 1992 Jun;67(6):1587-98. 6. Federici A, Bellagamba S, Rocchi MBL. Does dance based-training improve balance in adult and young old subjects? A pilot randomized controlled trial. Aging Clin Exp Res. 2005 Oct;17(5):385-89. 7. Lephart SM, Fu FH. Proprioception and Neuromuscular Control in Joint Stability. Champaign, IL: Human Kinetics, 2000. 8. Roberts TD. Understanding Balance: The Mechanisms of Posture and Locomotion. London, UK: Chapman & Hall, 2005. 9. Shumway-Cook A, Woolacott MH. Motor control: Theory and Practical Applications (2nd ed). Baltimore: Lippincott, Williams & Wilkins, 2001. 10. Martini FH. Fundamentals of Anatomy and Physiology (5th ed). Upper Saddle River, NJ: Prentice Hall, 2001. 11. Shiffman HR. Sensation and Perception: An Integrated Approach (5th ed). New York: John Wiley and Sons, 2001. 12. Golomer E, Crémieux J, Dupui P, et al. Visual contribution to selfinduced body sway frequencies and visual perception of male professional dancers. Neurosci Lett. 1999 Jun 4;267(3):189-92. 13. Golomer E, Dupui P. Spectral analysis of adult dancers’ sways: sex and inter-action vision-proprioception. Int J Neurosci. 2000 Nov;105(14):15-26. 14. Simmons RW. Sensory organization determinants of postural stability in trained ballet dancers. Int J Neurosci. 2005 Jan;115(1):87-97. 15. Perrin P, Deviterne D, Hugel F, Perrot C. Judo better than dance,
Journal of Dance Medicine & Science • Volume 18, Number 1, 2014 develops sensorimotor adaptablities involved in balance control. Gait Posture. 2002 Apr;15(2):187-94. 16. Huxham FE, Goldie PA, Patla AE. Theoretical considerations in balance assessment. Aust J Physiother. 2001;47(2):89-100. 17. Willardson JM. The effectiveness of resistance exercises performed on unstable equipment. Strength Cond J. 2004;26(5):70-4. 18. Gamble P. Strength and Conditioning for Team Sports: Sports Specific Physical Preparation for High Performance. Abingdon, Oxon: Routledge, 2010. 19. Redding E, Weller P, Ehrenberg S, et al. The development of a high intensity dance performance fitness test. J Dance Med Sci. 2009;13(1):3-9. 20. Wyon M, Redding E, Abt G, et al. Development, reliability, and validity of a dance specific aerobic fitness test (DAFT). J Dance Med Sci. 2003;7(3): 80-4. 21. Earl J, Hertel J. Lower-extremity muscle activation during the Star Excursion Balance Tests. J Sport Rehabil. 2001;10(2):93-104. 22. Hertel J, Miller S, Denegar C. Intratester and intertester reliability during the Star Excursion Balance Test. J Sport Rehabil. 2000;9(2):104-16. 23. Kinzey S, Armstrong C. The reliability of the Star-Excursion Balance Test in assessing dynamic balance. J Orthop Sport Phys Ther. 1998 May;27(5):356-60. 24. Lanning CL, Uhl TL, Ingram CL, et al. Baseline values of trunk endurance and hip strength in collegiate
athletes. J Athl Train. 2006 OctDec;41(4):427-34. 25. Batson G. Validating a dance-specific screening test for balance: preliminary results from multisite testing. Med Probl Perform Art. 2010 Sep;25(3):110-15. 26. Batson G, Tanaka ML, Long BL, et al. Performance strategies in collegeage dancers on the Modified Star Excursion Balance Test: preliminary quantitative and qualitative findings. In: Solomon R, Solomon J (eds): The 20th Annual Meeting of the International Association for Dance Medicine & Science. Birmingham, UK: IADMS, 2010, pp. 51-52. 27. Emery CA, Cassidy JD, Klassen TP, et al. Effectiveness of a homebased balance-training program in reducing sports related injuries among healthy adolescents: a cluster randomized controlled trial. CMAJ. 2005 Mar 15;172(6):749-54. 28. Kollmitzer J, Ebenbichler GR, Sabo A, et al. Effects of back extensor strength training versus balance training on postural control. Med Sci Sport Exerc. 2000 Oct;32(10):17706. 29. McGuine TA, Keene JS. The effect of a balance training program on the risk of ankle sprains in high school athletes. Am J Sports Med. 2006 Jul;34(7):1103-11. 30. Myer GD, Ford KR, Brent JL, Hewett TE. The effects of plyometric vs. dynamic stabilization and balance training on power, balance, and landing force in female ath-
11
letes. J Strength Cond Res. 2006 May;20(2):346-53. 31. Wildmering V, Krasnow D. Motor learning and teaching dance. International Association for Dance Medicine and Science. Retrieved from http://www.iadms.org/associations/2991/files/info/motor_learning.pdf. 2009. 32. Horak FB, Nutt JG, Nashner LM. Postural inflexibility in Parkinson’s subjects. J Neurolo Sci. 1992 Aug;111(1):46-58. 33. Hoffman M, Schrader J, Applegate T, Koceja D. Unilateral postural control of the functionally dominant and nondominant extremities of healthy subjects. J Athl Train. 1998 Oct;33(4):319-22. 34. van Emmerick REA, van Wegen EEH. On variability and stability in human movement. J Appl Biomech. 2000;16:394-406. 35. Golomer E, Dupui P, Monod H. The effects of maturation on self-induced dynamic body sway frequencies of girls performing acrobatics or classical ballet. Eur J Appl Occup Physiol. 1997;76(2):140-4. 36. Champion LM, Chatfield S. Measurement of turnout in dance research: a critical review, J Dance Med Sci. 2008;12(4):121-33. 37. Seber S, Haser B, Köse N, et al. Rotational profile of the lower extremity and foot progression angle: computerized tomographic examination of 50 male adults. Arch Orthop Trauma Surg. 2000;120(5-6):255-8.
Visual conditions for a dancer vary greatly between theatrical performance environments and dance studios, and this variability may be detrimental to their dynamic balance performance, particularly under stage lighting. In order to maintain balance control, dancers reportedly rely heavily on visual input, yet those who rely more on proprioceptive strategies for balancing have been found to be more stable. The purpose of this study was to assess the capability of an eyes-closed, dancespecific training program to nurture in dancers proprioceptive mechanisms that may facilitate their dynamic balance control. Eighteen elite pre-professional ballet dancers were randomly assigned to either a control (eyes open) or experimental (eyes closed) group for the intervention. The balance abilities of all subjects were tested using five dance-specific variations of the Star Excursion Balance Test before and after a 4 week balance intervention. Reach distance and time to complete the tests were recorded separately as indirect measurements of dynamic balance. The intervention consisted of dance-specific, eyes-closed exercises integrated into the dancers’ daily ballet class and designed progressively to challenge the dancers’ balance. During the intervention period, the control group undertook the same exercise program with their eyes open. Results revealed significant improvements in time to complete the three “timed” balance
tests, and distances reached significantly improved in one of the two “reach” balance tests. No significant improvements were observed in the control group for any variation of the tests. These results indicate that dancers can be trained to adopt proprioceptive strategies to maintain dynamic balance, which consequently improves their balance performance. Such findings could encourage use of eyes-closed training in daily dance classes due to its potential to improve dancers’ balance control.
C
lassical ballet has attracted audiences worldwide since the 17th Century.1 It is an art form that becomes more accessible when its performers have mastered the fundamental skills required to execute complex movement patterns. One such skill is balance. Balance is not only essential for successful ballet performance,2 but many classical choreographies showcase a dancer’s balance ability to excite an audience—for example, Princess Aurora in Sleeping Beauty. Classical ballet dancers must have sophisticated balance mechanisms in order to position themselves effectively during the complex choreographed sequences of dance practice and performance. Dance involves multidi-
Kimberley Hutt, M.Sc., and Emma Redding, Ph.D., are at the Trinity Laban Conservatoire of Music and Dance, London, United Kingdom. Correspondence: Kimberley Hutt, M.Sc., 42 Laurel Avenue, Potters Bar, Hertfordshire EN6 2AB, United Kingdom; [email protected]. Copyright © 2014 J. Michael Ryan Publishing, Inc. http://dx.doi.org/10.12678/1089-313X.18.1.3
rectional and rotational activities that challenge balance control, and dancers continually strive to improve balance in order to enhance their stability during the execution of a step or piece of choreography. Yim-Chiplis and Talbot3 stated that balance provides the foundation for human mobility and functional independence, highlighting the fundamental importance of balance to normal human locomotion. Research has shown that dance practice itself can autonomously provide effective balance training. 4 While the precise neurophysiological mechanism behind this effect is yet to be fully established, dancers have been shown to be exceptionally sensitive to variations in their gravitational axis.5 Consequently, research has used dance as a method to improve the balance ability of non-dancers6; however, dancers’ balance abilities need to be superior to those of the general population for successful and efficient dance performance. Balance refers to the ability to maintain the body’s center of gravity over its base of support.7 During dance, when the center of gravity is displaced beyond the vertical projection of the base of support, the dancer will become unstable and begin to fall.8 This explains why dynamic balance becomes more difficult as the base of support is reduced in size, for example, from fifth position to arabesque, from a flat foot to demi 3
4
Volume 18, Number 1, 2014 • Journal of Dance Medicine & Science
pointe, or en pointe as seen in classical ballet. To maintain balance, there must be an efficient and complex integration of the visual, vestibular, and somatosensory organs.9 Within these organs, there are specialized sensory receptors which provide information about the external environment (exteroceptors), maintain physiological homeostasis (interoceptors), and report on the position and movement of muscles and joints (proprioceptors).10 Proprioceptors within muscles and tendons are not only responsible for detecting body movement and position but also contribute to the control of postural reflexes to maintain stability while moving11 and are paramount in the prevention of injury due to their role in joint and postural stability. It seems prudent, therefore, to increase the challenge to these specialized receptors during dance practice with the aim of enhancing dancers’ proprioception and consequent balance abilities and potentially to reduce the risk of injury associated with loss of stability. There is general consensus in the literature that ballet dancers use proprioceptive and visual stimuli as their fundamental sensory inputs to maintain balance. Although as yet only anecdotal, research has suggested that due to the habitual use of mirrors during ballet class and rehearsals dancers rely too heavily on visual cues for dynamic balance while dancing.12 This could become problematic to a dancer’s balance when those visual cues change under different dance conditions. Based on reports of dancers’ dependence on visual input for balance control and when considering that visual conditions in theatrical performance settings are not representative of those in dance studios, the unfamiliar and unaccommodating visual conditions encountered in a theater may prove detrimental to dynamic balance during a dancer’s performance.2 Research has reported an ability for ballet dancers to modify the sensory organization that is responsible for balance control.13,14 This modification is known as a “shift,” and it refers to the transition from using one balance
mechanism to another. For example, dancers may use visual input as the dominant mechanism to maintain balance when their eyes are open, yet acutely shift to more proprioceptive strategies when their eyes are closed. A similar shift might be required under stage lighting conditions. As previously noted, classical ballet dancers’ balance abilities are reportedly superior to non-dancers’, yet when their eyes are closed, their balance abilities are no better,2 suggesting that their ability to shift acutely from one balance mechanism to another is unsophisticated. This could be explained by Hugel’s hypothesis that the practice of dance alone does not enhance the brain’s afferent signals of vestibular and proprioceptive stimulus.2 This in turn may be explained by the heavy reliance on visual stimulus and stable environments seen in dance practice. Perrin and coworkers15 found that Judokas have superior balance abilities compared to classical ballet dancers when tested with their eyes closed. The investigators inferred that Judo relies more heavily on proprioception than visual references, which may be advantageous to balance control. Therefore, research is needed to investigate the effects of balance training that favors proprioception as the dominant balance mechanism, with the aim of producing a chronic shift in sensory organization from an external visual dependency to a more proprioceptive “internal awareness” of stability when dancing. In their consideration of balance training for ballet dancers, Huxham, Goldie, and Patla16 stated that balance “cannot be separated from the action of which it is an integral component, or from the environment in which it is performed.” Willardson17 supported this theory, stating that balance is skill specific, and that it should be trained using the same technical skills that are required during performance. Therefore, in training dancers it seems appropriate to select a balance-training program with specificity to dance practice. The principle of specificity in sport is well documented and acknowl-
edged. 18 Considering the unique nature of dance and the purported benefits of specificity, it is no wonder that in the growing field of dance science researchers are developing dance-specific testing protocols and training regimens for a number of physiological, biomechanical, and aesthetic components of dance.19,20 Balance and stability testing must also be specific to the skills of the performer. The Star Excursion Balance Test (SEBT) has been said to stress dynamic balance control21-23 and is being used among the athletic population.24 The SEBT is arguably, therefore, more appropriate for testing dancers’ balance than other validated static balance tests. On the other hand, while the SEBT involves reaching a leg in multiple directions that replicate those seen in dance practice (for example, croisé and ouvert), it arguably lacks skill specificity to dance practice. In fact, while balance tests such as the SEBT are routinely implemented in screening and rehabilitation programs among the athletic and dance population, until recently no balance test has been designed with specificity to a given sport or activity, such as dance. The SEBT has now been modified (mSEBT) to improve its specificity to dance practice,25 and since the creation and initial testing of the mSEBT, researchers Glenna Batson and Margaret Wilson have been working collaboratively to improve the test.26 Developments to date have included randomization of the direction to which the working leg extends, in order to challenge the attentional focus of the dancer. As with the mSEBT, the “randomized” Star Excursion Balance Test (rSEBT) records measurements of reach distance and timings. While the randomized modifications have yet to be validated and published, the present investigators consider these further modifications valuable for their specificity to dance practice, due to the rapid responses to directional cues required when performing complicated dance sequences. The objective of the present study was to improve the dynamic balance
Journal of Dance Medicine & Science • Volume 18, Number 1, 2014
performance of a sample of dancers by increasing the challenge to their proprioceptive mechanisms of balance control in their daily dance practice. It was anticipated that the increased challenge could be achieved by integrating an eyes-closed training program to “switch off” the visual mechanisms, thus instigating a chronic shift from visual reliance to proprioceptive strategies for the maintainance of dynamic balance. The researchers hypothesized that incorporating eyesclosed practice into daily ballet classes would enhance compliance with the program and reduce the necessity for additional training. Significant improvements in dymanic balance abilites would provide sound rationale for dance educators to promote eyes-closed training in dancers’ daily practice.
Materials and Methods
Subjects The selection of subjects and protocol of the study were ethically approved prior to data collection and the intervention. All hard and electronic copies of subjects’ data were stored in compliance with the Data Protection Act of 1998, and anonymity was maintained throughout the study. Nineteen elite female pre-professional dancers from Central School of Ballet (UK) volunteered and consented to participate. The sample comprised the first year group, who had all been studying full time at the school for 9 months. Prior to testing, the researcher documented the subjects’ heights using a Bodymeter Measuring Tape (Factory Direct, USA) and recorded their mass in Newtons, taken from a portable force plate (Kistler®, Switzerland, threshold Fz < 10 mN, model 9286AA) and converted to kilograms in order to calculate BMI. For the participants’ epidemiology see
Table 1. Subjects were randomly assigned to one of two groups, control or experimental, by blindly selecting a piece of paper marked with “C” (control) or “E” (experimental) from a “hat.” This resulted in 10 experimental participants and 9 controls. At the beginning of the study, all subjects met the inclusion criteria, which stated that they should have no vestibular disorder or recent history of lower limb injury. During the intervention period, one dancer in the control group sustained an injury; therefore, data were collected for 10 dancers in the experimental group and 8 in the control group. Instrumentation A star-shaped grid, seen in Figure 1, was constructed using Cramer Tape Heavy Zinc Oxide (2.5 cm wide). The portable force plate was mounted longitudinally in the center of the grid. This was connected to a Kistler® DAQ-system which transmitted information of forces to a laptop computer equipped with Bioware System software version 4.1. Methodology The researcher determined leg dominance by asking the subjects to declare their favored “working leg” when dancing. Working leg was described to participants as being that which they preferred to “practice exercises with,” as opposed to “stand on.” Leg length data were collected using measurements from the greater trochanter to the distal lateral malleoli, in order that measurements of reach could be normalized. Testing procedures took place 3 days before and 3 days after a 4 week intervention. Procedures were explained to each subject, following which five variations of the original
5
Figure 1 The eight spoke grid and directions used for all variations of the SEBT. NB: These directions are based on the subject facing the front spoke and using her right leg as the dominant “reaching” leg.
SEBT were performed by each participant using their non-dominant leg as the supporting leg and their dominant leg as the reaching leg. The same researcher undertook all explanations and data collection to enhance intra-test reliability. The Basic SEBT Subjects were instructed to stand on the force plate with their supporting leg in parallel position and their dominant leg held in a parallel coup de pied (the “ready position,”25 Fig. 2). They were then instructed to bend the supporting leg and reach toward each spoke, first in a clockwise, then counter-clockwise direction, lightly “dabbing” the toe at the furthest point they could reach. Distance reached, missed spoke, falls, and near falls were recorded on a testing sheet. The entire test was terminated if a participant hopped or fell due to loss of balance or put excessive weight on the working foot when it was “dabbed” at each spoke. Subsequent Variations of the Basic SEBT On completion of the basic SEBT, the dancers were given instructions
Table 1 Participant Characteristics at the Start of the Study Mean Height (± SD) (cm)
Mean Mass (± SD) (kg)
Mean BMI (± SD)
Mean Leg Length (± SD) (cm)
Mean Age (± SD) (yrs)
Mean Years Dancing (± SD) (yrs)
163.82
50.75
18.95
82.26
16.68
11.16
(5.55)
(4.5)
(1.2)
(4.75)
(0.48)
(3.11)
6
Volume 18, Number 1, 2014 • Journal of Dance Medicine & Science
Figure 2 The “ready” position.
on the variations of four subsequent tests and performed the following modifications: 1. Modified SEBT—timed test (mSEBTtimed): The basic SEBT was performed as quickly and safely as possible. Participants were instructed to reach as far as possible in the fastest time they could. The time taken to finish one full circuit of the test in a clockwise and counter-clockwise direction was recorded, along with missed spokes, near falls, and falls. 2. Modified SEBT—timed test with cognitive interference (mSEBTcogint): The mSEBTtimed was performed while the researcher asked a series of questions for cognitive interference; for example, “Spell your surname backwards” and “What is one hundred minus thirteen?” Dancers with English as a second language were invited to answer the questions in their own language. Again, time taken to finish one full circuit of the test in a clockwise and counterclockwise direction was recorded, along with missed spokes, near falls, and falls. 3. Randomized SEBT—reach distance test (rSEBTreach): To stan-
dardize the test between subjects, an identical “randomized” order of the direction toward which the subject should reach was used by the researchers. Subjects were instructed to reach as far as they safely could, at their own pace. The subjects were taught the names given for the directions for each spoke (Fig. 1) and told that the direction to which they should reach would be given as soon as the working foot came back to the “ready position.” Wilson recommended the names of the spokes in research using the rSEBT (unpublished). It was considered that these terms would be easier for the dancer to interpret than the scientific terms used in standard research (e.g., anterior, posterior, medial, and lateral). Distance reached, missed spoke, falls, and near falls were recorded. 4. Randomized SEBT—timed test (rSEBTtimed): The rSEBTreach was repeated as fast and as safely as possible. For this variation of the test subjects were instructed to extend their leg to approximately 45°, rather than dab their toe to the floor. The researcher recorded time taken to complete the sequence, along with missed spokes, near falls, and falls. Due to suggestions made by Batson25 for future studies to observe quantitative force plate analysis, measurements of sway around the center of pressure (COP) in anteriorposterior (AP) and medial-lateral (ML) direction were recorded during all variations of the test. The present study did not obtain data using the foam modification of the mSEBT used by Batson,25 as the force plate would not transmit useful data of forces through foam. The Intervention Prior to the study, the lead researcher met with the participants’ regular ballet teacher to discuss the balance exercises that were to be integrated into the dancers’ daily class, forming the basis of the intervention. The researcher explained that the interven-
tion should be the first center practice exercises following barre work for the sake of consistency and should be performed 5 days per week for 4 weeks. The sequences were to become progressively challenging on a weekly basis, without posing any unnecessary risk to the dancers. The researcher suggested examples of such progressions, including stationary sequences progressing to travelling and turning sequences; double leg stance exercises progressing to single leg stance exercises; and sequences involving working legs a terre progressing to sequences with the working leg en l’air. When the dance-specific exercises had been choreographed by the dance teacher and approved by the researcher, the experimental group was instructed to perform the exercises with their eyes closed throughout the intervention period. They should only open their eyes during the eyes-closed sequence if they felt they were going to fall, and in that event, they should close their eyes and continue as soon as they felt they had regained their balance sufficiently to continue safely with the sequence. Throughout the intervention period the control group undertook the same program with their eyes open, as they would under normal dance class conditions. Data Analyses Mean (± SD) measurements were calculated for distance reached (cm), time to completion (s), or sway (mm) around the center of pressure (COP), depending on the balance test being performed. Using SPSS software (version 19; IBM, USA), with alpha set to 0.05, paired samples t-tests were carried out to assess for balance improvements following the intervention. Due to the use of multiple t-tests, Benjamini Hochberg corrections were undertaken as a valid means of reducing the likelihood of creating a type 1 statistical error, following which all p values were divided by two (alpha level = 0.025) to reflect the directional nature of the hypotheses—that reach distance and time would improve. The statistical analyses used mean data taken from each of the following
Journal of Dance Medicine & Science • Volume 18, Number 1, 2014
dependent variables: 1. Basic SEBT (reach distance measured in cm), 2. mSEBTtimed (time measured in seconds), 3. mSEBTcogint (time measured in seconds), 4. rSEBTreach (reach distance measured in cm), 5. rSEBTtimed (time measured in seconds), 6. Center of pressure – anteriorposterior (sway measured in mm), and 7. Center of pressure – mediallateral (sway measured in mm). Center of pressure (COP) measurements were only analyzed from data collected during the rSEBTtimed, as this was the only variation in which the working foot did not touch the floor. The act of “dabbing” the floor during all other variations of the test rendered force plate data unusable, due to extreme inter-subject variability of pressure with which the foot was “dabbed.”
Results Both groups appeared to improve in distance reached and time to completion of the balance tests; however, the improvements were greater among the experimental participants. Statistical analysis using paired samples t-tests revealed significant improvements in the balance of the experimental group
when undertaking four of the five variations of the test: basic SEBT (t [9] = -3.256, p = 0.01), mSEBTtimed (t [9] = 3.086, p = 0.01), mSEBTcogint (t = [9] 3.458, p = 0.01), and rSEBTtimed (t [9] = 4.643, p = 0.00). Conversely, no statistically significant differences were observed in the control group when undertaking any of the variations of the balance tests (Table 2). The control group demonstrated a 0.1% deterioration (increase) in time to complete the mSEBTtimed, an 8% improvement (decrease) in time to complete the mSEBTcogint, and a 4% improvement in time to complete the rSEBTtimed following the intervention period. More impressively, the experimental group demonstrated a 14% improvement in time to complete the mSEBTtimed, a 15% improvement in time to complete the mSEBTcogint, and a 16% improvement in time to complete the rSEBTtimed test (Fig. 3). When observing reach distances achieved, the control group managed to reach an average of 4% further during the basic SEBT and 0.3% further during the rSEBTreach after the 4-week period. These results proved to be non-significant. In the experimental group, the intervention elicited a significant increase of 5% reach distance achieved during the basic SEBT, but only a 0.9% increase during the rSEBTreach, which proved
7
to be non-significant (Table 2, Fig. 4). Neither trend nor significant difference was observed in any of the center of pressure measurements tested (control group: anterior-posterior – t [5] = -1.478, p = 0.34; medial-lateral – t [5] = 1.172, p = 0.19. Experimental group: anterior-posterior – t [7] = 1.537, p = 0.12; medial-lateral – t [7] = -0.825, p = 0.22). Therefore, it cannot be said that center of pressure measurements differed following an intervention to improve balance in dancers. In summary, results suggest that due to the observed trends in and largely significant improvements in the experimental group that were not present in the control group, the dynamic balance control of elite pre-professional ballet students did improve following a period of eyesclosed dance-specific training.
Discussion The purpose of the present study was to determine whether a 4-week, eyesclosed, dance-specific training program would improve dynamic balance in a sample of elite pre-professional dancers. To date, no other published research has reported the effectiveness of an eyes-closed training program on balance abilities in dancers. The dominant statistical findings revealed that when testing the dynamic balance ability of dancers performing
Table 2 Timings and Reach Distance for All Variations of the SEBT Basic SEBT (cm)
mSEBTtimed (s)
mSEBTcogint (s)
rSEBTreach (cm)
rSEBTtimed (s)
Pre
Post
Pre
Post
Pre
Post
Pre
Post
Pre
Post
Control
Mean (± SD)
69.9 (± 3.5)
72.4 (± 4)
16.5 (± 3.1)
16.5 (± 3)
18.1 (± 4.3)
16.6 (± 3.6)
74.3 (± 5.6)
74.5 (± 3.5)
29.2 (± 3.4)
28.1 (± 2.8)
Experimental Mean (± SD)
79.4 (± 5.6)
83.4 (± 7)
17.9 (± 3.4)
15.5 (± 3)
18.1 (± 3.3)
15.5 (± 3.1)
82.6 (± 6.6)
83.4 (± 6.5)
29.2 (± 3.1)
24.7 (± 2.3)
Descriptives
Paired samples t-test Control *Significance T value
NS (p = 0.07) -3.049
NS (p = 0.49) -.038
NS (p = 0.28) 1.251
NS (p = 0.51) -.161
NS (p = 0.24) 1.182
Experimental *Significance T value
SIG (p = 0.01) -3.256
SIG (p = 0.01) 3.086
SIG (p = 0.01) 3.458
NS (p = 0.22) 0.857
SIG (p = 0.00) 4.643
*NS = no significance; SIG = significant
8
Volume 18, Number 1, 2014 • Journal of Dance Medicine & Science
Figure 3 Mean times for both groups’ timed tests.
Figure 4 Mean reach distance for both groups’ reach tests.
variations of the mSEBT and rSEBT, time to complete all variations of the balance tests significantly improved following the intervention. Reach distances significantly improved when performing the basic SEBT but not during the rSEBTreach. There were no significant improvements observed among the control group across all variations of the balance tests. Theorizing that the test variations collectively represent dynamic balance, overall these results suggest that a 4-week, eyes-closed, dance-specific training program can improve the dynamic balance ability of dance students. Reach distance and timing may not be considered direct measures of balance. However, tasks such as these create perturbations that increase the challenge to the central nervous
system to generate appropriate motor responses. Indeed, Huxham, Goldie, and Patla 16 stated that increasing the speed of a moving limb requires greater equilibrium control to maintain balance control, thus justifying the inclusion of a test of speed in the mSEBT, rSEBT, and the present study. Interestingly, all variations of the timed tests performed in the present study revealed significant improvements; therefore, it may have been the dancers’ equilibrium control component of balance that improved to the greatest extent. The reach component of the basic SEBT and rSEBTreach involves changes in the magnitude, direction, and combination of forces acting on and produced by the body. As the direction of the reach was constantly
changing, the biomechanical challenges to maintain balance control were increased. The current study found inconclusive results when considering reach as an isolated measure of balance, with only the basic SEBT variation of the reach tests (and not the rSEBTreach) revealing significant improvements. However, the additional challenge produced by randomization involved in the rSEBTreach may have caused the more variable results. These additional challenges could be explained by Huxham, Goldie, and Patla’s16 description of two mechanisms, proactive and reactive, by which environmental factors influence the biomechanical properties of balance. Proactive balance control involves information received through the eyes but also includes a form of
Journal of Dance Medicine & Science • Volume 18, Number 1, 2014
control known as predictive control. Predictive balance control relies on anticipatory postural adjustments in response to internal representations of the body and learned movement.16 In the case of predictive control, muscle activity begins before the actual movement takes place, with the magnitude of these muscle responses being dependent upon the direction and speed of the movement. The randomization of direction used in the rSEBTreach variation would have certainly disrupted anticipatory mechanisms, as the anticipatory muscle activity could not have been able to respond in anticipation of a predicted direction, a factor that could account for the non-significant improvements when performing the rSEBTreach. Despite the lack of significance, it is noteworthy that the experimental group achieved greater improvements in reach distances for the rSEBTreach than the control group, and it may only have been a matter of time before significance was achieved. Testing durations of balance studies among healthy adolescents have ranged from 4 to 7 weeks,27-30 suggesting that 4 weeks of training, as used in the present study, is sufficient time to stimulate an effect. Many of these studies reported non-compliance to the training program as a limitation of the intervention. Dancers’ daily training loads are high, and interventions that impose exercises in addition to daily practice may not be adhered to. The present study boasts excellent compliance, as the exercises were incorporated into the dancers’ morning ballet class 5 days per week for the duration of the study. Compliance is likely to be greater when the training can be incorporated into the daily training activities, as ancillary practice will rely heavily on individual dedication to the research program. Batson25 reported variable execution of the balance tests during her preliminary study; however, her study’s protocol dictated that testers should not demonstrate the test to the subjects, but rather the testers instructed the subjects by reciting from a learned script. The researchers of
the present study saw this as a limitation to Batson’s study that may have resulted in the observed inter-subject variability and consequently chose to demonstrate the balance tests to the subjects prior to data collection. Perhaps as a result of this, the execution of the tests was generally consistent among dancers. Interestingly, the greatest variability in test execution was observed among dancers who had 5 years or less dance experience. Wilmerding and Krasnow31 state that perception (the dancer’s observation and attention to a given demonstration) and execution (the dancer’s replication of the given demonstration) are essential stages involved in motor learning. These stages are likely to be less refined in dancers with fewer years of experience. Such concepts are supported by the variability of test execution seen by Batson,25 who studied dancers with a mean of only 2.6 (± 2.7) years training in participants from the UK and 9.6 (± 4.6) years training in those from the USA. However, further research would be required to confirm a relationship between variability of execution of balance tests and years of dance experience. Following suggestions for future research by Batson,25 the present study analyzed force plate data to investigate postural sway as represented by COP measurements. No trends were observed in this way, and inter- and intra-participant results elicited considerable variability. One explanation for this may be “exploratory activity” in response to a new movement pattern. However, such activity is not necessarily destablizing 32; indeed, the notion of increased sway being a reflection of instability is a common misconception, and postural sway may not be a direct indicator of postural stability.33 Alternatively, large sway may be due to deliberate actions taken to maintain balance, rather than a measure of instability.32 van Emmerick and van Wegen34 suggest that increased oscillations identified by force plate data may be an indication of greater dynamic neuromuscular function to reduce instability. The
9
investigators deduce that increased postural dynamics enhance the ability of the body to adapt to imposed postural disturbances. While the precise neurophysiological adaptations behind the balance improvements observed in the present study have not been identified, it could be postulated that they were due to a trained (chronic) shift in sensory organization from visual to proprioceptive dependency to maintain balance control. Future studies might observe EMG activity to analyze muscular activity in response to the potential shift in sensory organization.
Study Limitations The present study did not assess the range of ankle dorsiflexion available to each participant, as did Batson.25 It is conceivable that several dancers reached the limit of dorsiflexion on their supporting leg before reach distance became sufficient to disturb their balance. In this case, it is unlikely that reach measurements would change following an intervention, as it would not be balance itself that was limiting the reach distance achievable but rather the range of motion available at the ankle. The present study determined leg dominance in an unorthodox manner, by asking participants what their favored “working leg” was when practicing dance. Traditional methods of determining leg dominance are varied, and there are disparities between “functional” (for example, the kicking leg in football) and “strength” (for example, the standing leg during postural control tasks) definitions of leg dominance.33 Failure to use an established method for determining leg dominance in the present study may have resulted in variability of test results among the participants. The entire sample of dancers used in the present study was female. When testing one gender it cannot be assumed that results are transferrable to the other, yet there is no literature to suggest that balance training affects genders differently. When testing dynamic balance Golomer35 observed that males are more dependent on vi-
10
Volume 18, Number 1, 2014 • Journal of Dance Medicine & Science
sion than females; therefore, research into the effects of eyes-closed training for dancers of both genders is merited. While the present study provides valuable information about balance training in female dancers in their late teens, both the dance and scientific literature generally contain considerable discussion of the variability in balance mechanisms employed at different ages. Therefore, the results from the present study cannot be generalized to dancers of all age groups, and future research might attempt to determine whether there is an optimal age for balance training. Furthermore, mechanisms of balance control between dancers with varied years of training should be investigated. While all dancers in the present study were considered to be “elite pre-professionals” and demonstrated similar abilities in dance classes, variability of execution of the balance tests was only observed in the dancers with fewer years of dance experience. All balance tests were performed in parallel position of the supporting leg, as prescribed by the original SEBT and subsequent modifications. The dancers in the present study were all aspiring ballet dancers; as such they favored a “turned out” position and felt uncomfortable in the required parallel position. In fact, for some ballet dancers, standing with the feet in parallel requires them to rotate their femurs internally. This is due to dancers commonly demonstrating greater tibial version angles than the general population36 and increased foot progression angles (or “out toeing”).37 To maintain specificity to dance, future studies could consider allowing ballettrained subjects to stand in turned out position, more closely to replicate balance abilities in their chosen dance style. Future researchers should take note that the data collected from the present study cannot be compared to that collected without the use of a force platform. The small elevated step produced by the force platform no doubt affected the distance the dancers were able to reach.
It is noteworthy that by chance there was a difference in balance abilities between the experimental and control groups pre-intervention, despite the random selection of participants. This difference proved to be significant when analyzed statistically. The researchers could not have foreseen this anomaly; however, the results should be interpreted carefully, as the control group may well have had more sophisticated balance strategies pretesting, which may have reduced any potential effects of the intervention exercises. Nonetheless, the exercises used for the intervention were not considered to be sufficiently challenging when performed with the eyes open (the control group condition) to elicit an improvement in balance in elite level dancers.
Conclusion To the best of the researchers’ knowledge, this study is the first to assess the effectiveness of eyes-closed training on the dynamic balance abilities of dancers. The research used speed and reach distance (of a lower limb) as indirect measurements of dynamic balance, and the resulting data indicate that eyes-closed training can improve the balance abilities of elite female pre-professional ballet dancers. These results imply that closing the eyes during dance training is an effective way to stimulate a shift from visual to proprioceptive dependency for balance control, thereby improving balance regardless of visual conditions in the surrounding environment. However, further research is needed to investigate the precise neural and biomechanical mechanisms by which the improvements observed in this study were achieved. These findings may then encourage dance educators and practitioners to explore the benefits to be derived from incorporating eyesclosed training into dance practice.
References 1. Kassing G. History of Dance: An Interactive Arts Approach. Champaign, IL: Human Kinetics Publishers, Inc., 2007. 2. Hugel F, Cadopi M, Kohler F, Perrin
P. Postural control of ballet dancers: a specific use of visual input for artistic purposes. Int J Sports Med. 1999 Feb;20(2):86-92. 3. Yim-Chiplis PK, Talbot LA. Defining and measuring balance in adults. Biol Res Nurs. 2000 Apr;1(4):32131. 4. Brown S, Martinez MJ, Parsons LM. The neural basis of human dance. Cereb Cortex. 2006 Aug;16(8):115767. 5. Mouchnino L, Ayrenty R, Massion J, Pedotti A. Coordination between equilibrium and head-trunk orientation during leg movement: a new strategy built up by training. J Neurophysiol. 1992 Jun;67(6):1587-98. 6. Federici A, Bellagamba S, Rocchi MBL. Does dance based-training improve balance in adult and young old subjects? A pilot randomized controlled trial. Aging Clin Exp Res. 2005 Oct;17(5):385-89. 7. Lephart SM, Fu FH. Proprioception and Neuromuscular Control in Joint Stability. Champaign, IL: Human Kinetics, 2000. 8. Roberts TD. Understanding Balance: The Mechanisms of Posture and Locomotion. London, UK: Chapman & Hall, 2005. 9. Shumway-Cook A, Woolacott MH. Motor control: Theory and Practical Applications (2nd ed). Baltimore: Lippincott, Williams & Wilkins, 2001. 10. Martini FH. Fundamentals of Anatomy and Physiology (5th ed). Upper Saddle River, NJ: Prentice Hall, 2001. 11. Shiffman HR. Sensation and Perception: An Integrated Approach (5th ed). New York: John Wiley and Sons, 2001. 12. Golomer E, Crémieux J, Dupui P, et al. Visual contribution to selfinduced body sway frequencies and visual perception of male professional dancers. Neurosci Lett. 1999 Jun 4;267(3):189-92. 13. Golomer E, Dupui P. Spectral analysis of adult dancers’ sways: sex and inter-action vision-proprioception. Int J Neurosci. 2000 Nov;105(14):15-26. 14. Simmons RW. Sensory organization determinants of postural stability in trained ballet dancers. Int J Neurosci. 2005 Jan;115(1):87-97. 15. Perrin P, Deviterne D, Hugel F, Perrot C. Judo better than dance,
Journal of Dance Medicine & Science • Volume 18, Number 1, 2014 develops sensorimotor adaptablities involved in balance control. Gait Posture. 2002 Apr;15(2):187-94. 16. Huxham FE, Goldie PA, Patla AE. Theoretical considerations in balance assessment. Aust J Physiother. 2001;47(2):89-100. 17. Willardson JM. The effectiveness of resistance exercises performed on unstable equipment. Strength Cond J. 2004;26(5):70-4. 18. Gamble P. Strength and Conditioning for Team Sports: Sports Specific Physical Preparation for High Performance. Abingdon, Oxon: Routledge, 2010. 19. Redding E, Weller P, Ehrenberg S, et al. The development of a high intensity dance performance fitness test. J Dance Med Sci. 2009;13(1):3-9. 20. Wyon M, Redding E, Abt G, et al. Development, reliability, and validity of a dance specific aerobic fitness test (DAFT). J Dance Med Sci. 2003;7(3): 80-4. 21. Earl J, Hertel J. Lower-extremity muscle activation during the Star Excursion Balance Tests. J Sport Rehabil. 2001;10(2):93-104. 22. Hertel J, Miller S, Denegar C. Intratester and intertester reliability during the Star Excursion Balance Test. J Sport Rehabil. 2000;9(2):104-16. 23. Kinzey S, Armstrong C. The reliability of the Star-Excursion Balance Test in assessing dynamic balance. J Orthop Sport Phys Ther. 1998 May;27(5):356-60. 24. Lanning CL, Uhl TL, Ingram CL, et al. Baseline values of trunk endurance and hip strength in collegiate
athletes. J Athl Train. 2006 OctDec;41(4):427-34. 25. Batson G. Validating a dance-specific screening test for balance: preliminary results from multisite testing. Med Probl Perform Art. 2010 Sep;25(3):110-15. 26. Batson G, Tanaka ML, Long BL, et al. Performance strategies in collegeage dancers on the Modified Star Excursion Balance Test: preliminary quantitative and qualitative findings. In: Solomon R, Solomon J (eds): The 20th Annual Meeting of the International Association for Dance Medicine & Science. Birmingham, UK: IADMS, 2010, pp. 51-52. 27. Emery CA, Cassidy JD, Klassen TP, et al. Effectiveness of a homebased balance-training program in reducing sports related injuries among healthy adolescents: a cluster randomized controlled trial. CMAJ. 2005 Mar 15;172(6):749-54. 28. Kollmitzer J, Ebenbichler GR, Sabo A, et al. Effects of back extensor strength training versus balance training on postural control. Med Sci Sport Exerc. 2000 Oct;32(10):17706. 29. McGuine TA, Keene JS. The effect of a balance training program on the risk of ankle sprains in high school athletes. Am J Sports Med. 2006 Jul;34(7):1103-11. 30. Myer GD, Ford KR, Brent JL, Hewett TE. The effects of plyometric vs. dynamic stabilization and balance training on power, balance, and landing force in female ath-
11
letes. J Strength Cond Res. 2006 May;20(2):346-53. 31. Wildmering V, Krasnow D. Motor learning and teaching dance. International Association for Dance Medicine and Science. Retrieved from http://www.iadms.org/associations/2991/files/info/motor_learning.pdf. 2009. 32. Horak FB, Nutt JG, Nashner LM. Postural inflexibility in Parkinson’s subjects. J Neurolo Sci. 1992 Aug;111(1):46-58. 33. Hoffman M, Schrader J, Applegate T, Koceja D. Unilateral postural control of the functionally dominant and nondominant extremities of healthy subjects. J Athl Train. 1998 Oct;33(4):319-22. 34. van Emmerick REA, van Wegen EEH. On variability and stability in human movement. J Appl Biomech. 2000;16:394-406. 35. Golomer E, Dupui P, Monod H. The effects of maturation on self-induced dynamic body sway frequencies of girls performing acrobatics or classical ballet. Eur J Appl Occup Physiol. 1997;76(2):140-4. 36. Champion LM, Chatfield S. Measurement of turnout in dance research: a critical review, J Dance Med Sci. 2008;12(4):121-33. 37. Seber S, Haser B, Köse N, et al. Rotational profile of the lower extremity and foot progression angle: computerized tomographic examination of 50 male adults. Arch Orthop Trauma Surg. 2000;120(5-6):255-8.
