- Email: [email protected]
The effect of external perturbations on variability in joint coupling and single joint variability
Human Movement Science xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Human Movement Science journal homepage: www.elsevier.com/locate/humov
The effect of external perturbations on variability in joint coupling and single joint variability Anita Haudum ⇑, Jürgen Birklbauer, Erich Müller Department of Sport Science and Kinesiology, University of Salzburg, Salzburg, Austria Doppler Laboratory ‘‘Biomechanics in Skiing’’, Rifer Schlossallee 49, 5400 Hallein/Rif, Salzburg, Austria
a r t i c l e
i n f o
Article history: Available online xxxx PsycINFO classification: 2330 2343 3720 Keywords: Single joint variability Coupling variability Perturbations Running
a b s t r a c t This paper explores the effect of goal-oriented external perturbations created by elastic tubes attached to the hip and ankles on lower limb joint variability and hip–knee and knee–ankle coordination variability during running. Kinematics of ten healthy male runners were analysed prior to and following a 7-week tube running intervention while running with and one without this constraint. The training intervention was based on variable training aspects to increase within-movement variability and adaptability of the running pattern. To analyse the effects of the tubes on the running pattern, the phase plot vector length deviation (i.e., the standard deviation of the phase plot vector length) for the within-joint variability and the continuous relative phase variability for the joint coupling variability were calculated. Results revealed acute increases of variability in both parameters. However, after the intervention, variability of the tube running situation returned to normal for all couplings and joints except the knee. No transfer effects to normal running were observed. This suggests very rapid adaptations to such perturbations. In the long-term, it may ask for more or different variations. Ó 2014 Elsevier B.V. All rights reserved.
⇑ Corresponding author. Address: Department of Sport Science and Kinesiology, University of Salzburg, Rifer Schlossallee 49, 5400 Hallein/Rif, Salzburg, Austria. Tel.: +43 664 7505 7177; fax: +43 662 60 44 614. E-mail address: [email protected] (A. Haudum). http://dx.doi.org/10.1016/j.humov.2014.02.004 0167-9457/Ó 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: Haudum, A., et al. The effect of external perturbations on variability in joint coupling and single joint variability. Human Movement Science (2014), http://dx.doi.org/10.1016/ j.humov.2014.02.004
2
A. Haudum et al. / Human Movement Science xxx (2014) xxx–xxx
1. Introduction Running has always been a very popular recreational activity. However, improved fitness and growing ambition lead to an increased amount and intensity of training that is often associated with overuse injuries in the hip, knee or ankle (Hamill, van Emmerik, Heiderscheit, & Li, 1999; Heiderscheit, Hamill, & van Emmerik, 2002; Hein et al., 2012; Stergiou, Jensen, Bates, Scholten, & Tzetzis, 2001). Due to pain-related movement, joint coordination variability has been shown to decrease as a kind of protective mechanism (Van Emmerik & Van Wegen, 2000). The loss of variability, however, results in reduced adaptability or flexibility, which are necessary for optimal functioning of healthy systems (Heiderscheit et al., 2002). As the mastering and organization of the many degrees of freedom allows the adoption of functional movement patterns (Bernstein, 1967), analysing changes due to continuous training in healthy individuals may reveal possible changes in joint coordination variability and help prevent injuries. Hence, a large number of running studies (e.g., Dierks & Davis, 2007; Gittoes & Wilson, 2010; Hamill et al., 1999; Heiderscheit et al., 2002; Hein et al., 2012) investigated either two populations (i.e., one healthy and one pathological) or report only single measurements done with respect to lower limb joint kinematics and coordination variability. However, limited research is available focusing on the application of stimuli intended to increase variability to regain or maintain a healthy amount of coordination or coupling variability. Since it is largely accepted that variability is a functional entity that helps establish optimal states of coordination (Hamill, Palmer, & Van Emmerik, 2012; Wheat & Glazier, 2006), investigations that apply perturbations to the movement pattern may help to understand whether inducing external goaldirected perturbations enables an increase of functional variability within the movement pattern (Haudum, Birklbauer, Kröll, & Müller, 2011, 2012). That is, through the application of additional variability, the movement pattern may again become more flexible and, therefore, more adaptable to external perturbations (Haudum et al., 2011, 2012; Schöllhorn, Hegen, & Davids, 2012; Schöllhorn, Mayer-Kress, Newell, & Michelbrink, 2009). This may further help to move more economically due to a transition to a new or at least slightly altered behavioral pattern (Haudum, Birklbauer, & Müller, 2011b). In running, where many degrees of freedom are orchestrated to a stable, coordinated and skillful movement, this approach may include the effect of running under changing conditions or responses to intervention programs in a healthy runners’ population (Stergiou, Jensen et al., 2001). An example for changing conditions or constraints can be obstacles (Stergiou, Jensen et al., 2001), type of shoes (Stöggl, Haudum, Birklbauer, Murrer, & Müller, 2010) or the application of elastic tubes (Haudum, Birklbauer, & Müller, 2012b; Haudum et al., 2011; Haudum et al., 2012), which may allow a more precise addition of variability. Several running studies are available on acute influences of varied conditions, where coordination has been examined such as obstacles in the track (Stergiou, Jensen et al., 2001; Stergiou, Scholten, Jensen, & Blanke, 2001) or elastic tubes applied to the lower extremities (e.g., Haudum et al., 2012). To our knowledge, only few intervention studies exist where joint coordination was examined especially under varying conditions (e.g., Haudum, Birklbauer, & Müller, 2011a; Haudum et al., 2012b). A limited number of studies, however, have employed intervention-induced variability to maintain or regain a functional level of variability in healthy individuals in joint coupling variability (Haudum et al., 2012b), which may provide relevant insights into possible alterations due to the prevailing constraints (Hein et al., 2012). In addition, the effect of variable constraints applied in acute situations or over a longer period of time may differ, since a new and possibly more appropriate movement pattern may be developed. In a first step and to clarify the role of a dynamic constraint in the form of elastic tubes on joint motion and joint coupling variability in a healthy population, we employed kinematic motion analyses to determine the effects of such an intervention. The elastic tubes were used to apply perturbations to the running pattern so that within movement variability is increased but the running patter is not rendered. The tubes are used to functionally achieve an optimal level of within-movement variability and to support the shaping or initiating of emergent movement patterns. Due to the sensitivity of the
Please cite this article in press as: Haudum, A., et al. The effect of external perturbations on variability in joint coupling and single joint variability. Human Movement Science (2014), http://dx.doi.org/10.1016/ j.humov.2014.02.004
A. Haudum et al. / Human Movement Science xxx (2014) xxx–xxx
3
initial conditions, the tubes have a different impact and, therefore, ask for permanent adaptations to their perturbations. Since the tubes demand a flexible movement pattern and permanent adaptations, we hypothesize that movement variability is influenced by the harness. Hence, the main objective of the study was to determine the influences of dynamic constraints (i.e., elastic tubes attached to the legs to slightly increase within-movement variability) prior to and following a 7-week training intervention on continuous relative phase variability (CRPV) and single joint variability (SJV) and whether there are differences in adaptation between CRPV and SJV. Since coordination is the functional link between muscles, joints and body segments to achieve a desired movement outcome (Tepavac & Field-Fote, 2001), the investigation of both may be of high importance. Since the magnitude of variability depends on the calculations being used, it need not be the case that all parameters reflect the same direction (Mullineaux, 2007). Further, as the results may differ when analysing the entire stride or when analysing stance or swing separately (Hamill et al. 1999), stride as well as stance and swing were analysed. We hypothesized that tube running will result in higher acute variability than normal running and that the initially higher tube running variability will decrease throughout the intervention due to tube running adaptation. We further hypothesize a transfer effect of the tube running intervention on normal running variability due to a generalization of the tube running pattern because of the proximity of the two running conditions (i.e., running with the tubes and running without tubes). With respect to the amount of influences of the variable constraint, we expected differences between entire stride, stance phase or swing calculations. 2. Methods 2.1. Participants Thirteen recreational runners gave their written informed consent to participate in the study. All runners had either participated in prior treadmill running interventions or were used to train on a treadmill and therefore considered as treadmill experienced. Of the original sample size, ten runners (mean age: 26.1 ± 7.1 years, mean height: 177.1 ± 7.2 cm, mean weight: 72.0 ± 6.6 kg) could be included in the analysis due to illness and injury of the 3 other participants. None of the runners had prior experience in running with the constraints (Fig. 1). The study was approved by the local ethics committee. 2.2. Training device A padded harness (Fig. 1; Tendybelt, Salzburg, Austria) was used to fix the tubes (Thera-BandÒ GmbH, Dornburg-Frickhofen, Germany) at both ankles (i.e., at the heel tab of the running shoes) and the lower back (ilio-sacral joint). Due to the results observed in pre-investigations, the tube length was set at 40% of the individual leg length (Haudum et al., 2012; Haudum, Birklbauer, & Müller, 2012a; Haudum et al., 2012b). The tubes are attached in order to apply variability to the running pattern, whereby the tubes’ resistance should in no way render running with the current running pattern impossible; rather, the tubes’ resistance should create additional variability within the movement skill. The functional role of the tubes is to achieve an optimal level of within-movement variability and to support the shaping or initiating of emergent movement patterns. The chosen tube length should ensure that small perturbations would increase within running variability and require permanent adaptations to the tube perturbations. 2.3. Training intervention Participants completed 18 training sessions of tube running on a treadmill (10.5 km.h 1 and 0% grade) over 7 weeks. Alternately, 2 and 3 sessions were run each week. The session duration was increased from 45 min for sessions 1–2 to 50 min for sessions 3–6 and to 55 min for sessions 7–18. The exercises during the training session were the same for each participant. Please cite this article in press as: Haudum, A., et al. The effect of external perturbations on variability in joint coupling and single joint variability. Human Movement Science (2014), http://dx.doi.org/10.1016/ j.humov.2014.02.004
4
A. Haudum et al. / Human Movement Science xxx (2014) xxx–xxx
Fig. 1. One participant running with the training device.
2.3.1. Training contents The aim of the training contents was to best possibly apply variability in the form of external perturbations within the movement pattern. The versatile exercises should help establish the individually most effective and optimal movement pattern. The intervention included different tube running exercises and was conducted according to differential learning (Schöllhorn et al., 2009) and training guidelines proposed by Birklbauer, Haudum, and Müller (2006). The differential learning approach (Schöllhorn et al., 2009) is considering movement variations as the basis of learning rather than movement repetitions and it is, therefore, taking advantage of the necessity of fluctuations or errors for learning (Schöllhorn, Hegen & Davids, 2012). These fluctuations through the versatile exercises represent movement differences that require constant reactions and adaptations to the varying constraints and which according to Schöllhorn provided essential information for refining or a self-organization of a better movement pattern. The guidelines (Birklbauer et al., 2006) were used to structure the exercises with respect to the amount of variability and where variability was induced. These guidelines emphasise on key points Please cite this article in press as: Haudum, A., et al. The effect of external perturbations on variability in joint coupling and single joint variability. Human Movement Science (2014), http://dx.doi.org/10.1016/ j.humov.2014.02.004
A. Haudum et al. / Human Movement Science xxx (2014) xxx–xxx
5
of the movement to be considered when applying variability and structure exercises (Haudum et al., 2012b). Examples of these guidelines were that the induced variability must be applied so that it supports the development of the movement, which means that the variability is intervention-induced applied; then the range of difference should decrease throughout the learning/intervention process. That is, a beginner needs larger differences between two consecutive exercises to perceive the difference between the two stimuli while an advanced runner or expert with the tubes can perceive even smaller differences between exercises due to his advanced perception of tube stimuli (i.e., larger differences at the beginning and smaller ones at the end of learning, which actually results in the same perceived differences) (Haudum et al., 2012b). This should help the runners utilize the tubes (i.e., the effect of reactive phenomena). The various exercises were chosen with respect to three key classes to provoke perturbations within the movement pattern. Each one of the three classes had three sub-aspects which were: (A) Intended variations with the participants focusing on the tubes: active work against the tubes; optimal work (i.e., use the tubes forces); supporting the tubes reactions (examples for such instructions were: running without completely stretching the tubes in any situation or bring the knee more upward and forward than usual); (B) Differences between the left and right leg in the position of the tubes: the tubes being attached between hip and shank; the tubes being attached between hip and the laces of the running shoe; no tubes attached; (C) Differences in the length of the tubes: shorter tubes so that running for a given time is at the utmost possible; the tube position as in the test runs (between hip and ankle); running without tubes. 2.4. Procedure Pre- and post-tests were held prior to and after a 7-week training intervention. Both times, 2 30 min tests (1 with and 1 without tubes) were run on a motorized treadmill (HP Cosmos Quasar 170/65, Traunstein, Germany). Both experimental running conditions were performed at 10.5 km.h 1 at 0% grade. This speed was chosen due to preliminary experiments (Haudum et al., 2012; Haudum et al., 2012a, 2012b) on the tubes behavior at different speeds and due to the participants’ endurance performance. Prior to the start of each test, runners were instructed to run without tubes for 5 min to warm up (8.5 km.h 1; 0% grade). Five runners started with the tubes, the other 5 with normal running. In the normal running situation, runners wore the harness but without the tubes. Runners were provided a 60-min recovery time between the two tests. Kinematics was recorded in all runs. 2.5. Data collection and processing Kinematic data were sampled in 2-min blocks starting at minutes 0, 3, 13, 16, 25 and 28. Of each 2min block the first 90 strides of recorded data were selected for further analysis, except in the first 2min block of each test run (i.e., min 0–2). Here, the first 10 strides were not included; however, the subsequent strides (i.e., 11–100) were selected for analysis to ensure that treadmill acceleration was finished and no influences due to different speed distorted the data. To collect kinematic data, an 8-camera Vicon three-dimensional motion analysis system (Vicon Peak, Oxford, UK) was used to analyse ankle, knee and hip movement in the sagittal plane. A total of 41 markers with a diameter of 14 mm were affixed to specific anatomical landmarks according to the Plug-In-Gait model (Plug-In-Gait Marker Set, Vicon Peak, Oxford, UK) (Haudum et al., 2012a, 2012b). The 3D-trajectories of all markers were analyzed utilizing Nexus software (Nexus 1.3, Vicon, Oxford, UK). Data were sampled at a rate of 250 Hz. The manually labelled marker trajectories were smoothed via a Woltring routine (MSE value of 10) (Woltring, 1986) and kinematics were calculated using the Plug-in-Gait model (Haudum et al., 2012b). Joint angles for the right leg ankle (plantar-dorsiflexion), knee (flexion–extension), and hip (flexion–extension) and angular velocities were exported to IKE-master (IKE-Software Solutions, Salzburg, Austria) for further analyses. Each stride cycle was defined from right heel contact to the next right heel contact. In addition, stance phase and swing were separately analysed. Heel contact and toe off were identified using the vertical velocities and the position profiles of the heel and toe markers as described by Fellin, Rose, Royer, and Davis (2010) and Lamoth, Daffertshofer, Huys, and Beek (2009), respectively. All angular Please cite this article in press as: Haudum, A., et al. The effect of external perturbations on variability in joint coupling and single joint variability. Human Movement Science (2014), http://dx.doi.org/10.1016/ j.humov.2014.02.004
6
A. Haudum et al. / Human Movement Science xxx (2014) xxx–xxx
data were interpolated to 101 data points and normalized to 0–100% of stride, stance phase or swing (Haudum et al., 2012b). 2.6. Data analysis Both intralimb coupling variability and single joint variability analyses were performed to determine movement variability of the right lower extremity. 2.6.1. Joint coupling variability The intralimb inter-joint coupling variability was assessed using continuous relative phase variability (CRPV). The coordination of the hip and knee joint and the knee and ankle joint was analysed according to the method of Hamill et al. (1999). That is, phase plots of the calculated joint angles were calculated with each phase plot consisting of the angle on the horizontal axis and the respective angular velocity on the vertical axis. In order to adjust the amplitude differences in the different joint angle (i.e., hip, knee, ankle joint) in the ranges of motion and centre the plots about the origin, angular displacement normalized to a range of 1 to +1. The angular velocity was normalized to the absolute maximum value to maintain zero velocity at the origin (Hamill, Haddad, & McDermott, 2000; Li, van den Bogert, Caldwell, van Emmerik, & Hamill, 1999). The normalization procedure was always conducted with respect to the maximum value over all strides within each 2-min block to maintain the true spatial properties. For all three joints, phase plots were employed to compare the two running situations with the phase angle being defined as the angle between the right horizontal and a line drawn from the origin to a specific data point. Subtracting the distal joint from the proximal, the hip knee and knee ankle continuous relative phases were calculated in the range of 0–360° avoiding discontinuities (Hamill et al., 2000; Wheat & Glazier, 2006). To determine CRPV, the angular deviation, which is the circular equivalent to the standard deviation in linear statistics, was calculated due to the directional nature of continuous relative phase (Batschelet, 1981; Miller, Chang, Baird, Van Emmerik, & Hamill, 2010). For each participant, the calculated individual CRPV was averaged across the complete stride, stance phase or swing for statistical analyses. 2.6.2. Single joint variability To separately analyse variability of each investigated joint, the phase plot vector length deviation (i.e., the dispersion of the vector lengths in the normalized phase plots) was applied. The variability for these linear data was calculated on a point-by-point basis across all 90 strides for the hip, knee and ankle joints in the form of the standard deviation. Similar to CRPV, the single joint variability (SJV) was averaged across the complete stride, stance phase or swing for statistical analyses. 2.7. Statistical analysis The IBM SPSS software Ver. 20.0. (SPSS Inc., Chicago, IL) was used for statistical analyses. For each participant, the mean joint coupling variability and the mean single joint variability was calculated in each running condition and for each 2-min block at each test (pre- and post-test). We performed 2 2 6 repeated measures ANOVAs (RMANOVA) with situation (tubes or normal running), time (pre- and post-test) and data blocks (2 min data block) as factors for both CRPV and SJV. 2 6 RMANOVA were calculated to estimate differences between the different pre- and post-test runs. Significant differences were reported at p < .05. Effect size partial eta squared (pg2) was additionally calculated. 3. Results 3.1. Continuous relative phase variability Looking at the group means of the averaged CRPV, which are graphically presented for stride (Fig. 2a and b), stance phase (Fig. 2c and d) and swing (Fig. 2e and f), the significant increases in Please cite this article in press as: Haudum, A., et al. The effect of external perturbations on variability in joint coupling and single joint variability. Human Movement Science (2014), http://dx.doi.org/10.1016/ j.humov.2014.02.004
A. Haudum et al. / Human Movement Science xxx (2014) xxx–xxx
7
Fig. 2. Average CRPV graphs for hip–knee (left column) and knee–ankle (right column) ±95% confidence interval. Graphs a and b represent averages for stride, graphs c and d averages for stance phase, and graphs e and f averages for swing. s = normal running pre-test; d = tube running pre-test; = normal running post-test; = tube running post-test.
CRP variability in the tube run ranged between 120% and 130% compared to the normal running level at the pre-test (p < .05; pg2 > .42). The CRPV was rather similar in both running conditions toward the end of the pre-test blocks. Separate analyses unveiled that the stride and stance phase data showed an increased variability for the hip–knee coupling in the tube running test at the pre-test which was reversed at the post test runs in comparison to normal running (p < .05; pg2 > .73). That is, the post-test unveiled approximately 15% less variability in the tube situation. While there was no significant result in the pre-test for the swing data, there was a considerable difference for the hip–knee CRPV in the post-test runs reasoned from the reduced tube running CRPV, which marginally failed significance (p < .06; pg2 > .55). For the knee–ankle complex, swing and stride data showed an initial increase to 120% of the normal running variability in the pre-test tube situation but apart from the initial increase data were Please cite this article in press as: Haudum, A., et al. The effect of external perturbations on variability in joint coupling and single joint variability. Human Movement Science (2014), http://dx.doi.org/10.1016/ j.humov.2014.02.004
8
A. Haudum et al. / Human Movement Science xxx (2014) xxx–xxx
rather similar for both tests which resulted in no significant differences (p > .05; pg2 < .15). The ankle– knee coupling variability during stance phase was reduced in both post-test runs compared to the pretest runs (p < .05; pg2 > .66). No further significant differences were found for this coupling.
3.2. Single joint variability Results for the SJV are displayed in Fig. 3 for stride (3a–c), stance phase (3d–f) and swing (3g–i). The variability increases due to the tubes were up to 170% for the knee compared to the normal running variability (p < .05; pg2 > .54) and for the hip (p < .05; pg2 > .56) and ankle up to 140% compared to normal running variability. The intervention-induced changes were very similar for all three, stride, stance phase and swing. That is, for both stance phase and swing as well as the total stride data, the increased variability in all three joints in the tube running pre-test was annihilated in the post-test runs (p < .05; pg2 > .45). For the knee joint, variability for normal running was significantly higher at post-test compared to the variability during tube running (p < .05; pg2 > .50). For both, SJV and CRPV, the significant 2 2 6 interaction is based on the initial increases in the tube running situation (p < .05; pg2 > .32). While the increases over several 2-min blocks in the pretest tube run, in the post-test tube run the initial 2-min blocks showed increased variability, which returned to normal rather fast. While the tube running variability changed from pre- to post-test, no transfer effects to normal running were found for either CRPV or SJV (p > .05; pg2 < .10).
Fig. 3. Average SJV graphs for hip, knee and ankle ±95% confidence interval. Graph a–c represent averages for stride, graphs d–f averages for stance phase, and g–h averages for swing. s = normal running pre-test; d = tube running pre-test; = normal running post-test; = tube running post-test.
Please cite this article in press as: Haudum, A., et al. The effect of external perturbations on variability in joint coupling and single joint variability. Human Movement Science (2014), http://dx.doi.org/10.1016/ j.humov.2014.02.004
A. Haudum et al. / Human Movement Science xxx (2014) xxx–xxx
9
4. Discussion The aim of this study was to investigate the effect of a varied training intervention on lower limb coordination variability (CRPV) and single joint variability (SJV) in healthy male runners. Using continuous relative phase and single joint analyses, we examined variability from two different perspectives. Based on our results and with the caution needed to compare different analyses techniques due to their different quantitative findings or requirements (Mullineaux, 2007, p. 107), an influence of the variability constraint was evident in almost all evaluated parameters, especially for the pre-test results. The magnitude of variability and of the constraints’ influence, however, differed between within-joint analyses and between-joint analyses. That is, the single joint approach exhibited a higher vulnerability (or stronger influence due) to the constraints than analyses between joints. It can be assumed that the within-joint variability represents a more flexible movement level while the betweenjoint parameters indicate that a robust movement path does exist (Haudum et al., 2012a, 2012b; Nigg, 2001). With respect to co-variation, which reflects the coupling between two joints, it can be stated that the movement pattern for the analysed skill seem well defined. Nevertheless, this difference between observation levels supports the selection of both analysis strategies as the amount of the actual influence varied between the methods. The tendencies of our data are in line with data from another locomotion intervention (i.e., walking), where another kind of variability constraint was used. In this intervention study on walking with unstable MBT shoes (Stöggl et al., 2010), the authors showed that after several weeks of walking with MBT shoes, which also intends to increase variability, repeated walking with MBT shoes also led to reduced variability during MBT walking. Both studies support the perfect adaption ability of the human system but they also demonstrate that it requires also new stimuli to keep the level of variability higher. Since the data showed slightly increased variability at the initial data block of the post-test tube run, different tube applications during the test runs might have revealed different effects on the SJV or CRPV. The increased variability during the first exposure supports results of other studies (Haudum et al., 2012; Haudum et al., 2012a). One explanation for the largely increased SJV and CRPV in the pre-tube runs is that the higher variability may serve an attempt to adjust the running pattern in the variability device-condition (Heiderscheit et al., 2002). Due to the observed decrease of variability especially in the SJV parameters, this strategy might be successful in the beginning as a kind of exploratory behavior to accommodate to the tubes (Haudum et al., 2012; Haudum et al., 2011a). However, while the strategy may be initially effective, tube running practice resulted in similar variability of both tube running and normal running after the intervention. Since normal running was not influenced by the intervention, a normalization of the tube running pattern occurred because the variability in the tube situation rather fast approached the variability level of the normal running situation (i.e., similar variability in normal running and tube running kinematics) and the assumption that such intervention leads to behavioral changes in the form of functional CRPV or SJV increases could not be confirmed. When examining CRPV averaged over the entire stride, stance phase or swing, results did not show significant differences, though tendencies towards different amounts of variability. The graphical illustrations (Fig. 4) demonstrated that despite the missing statistically significant differences between the two situations, the temporal structure or the variability peaks differed (Hein et al., 2012; Mullineaux, 2007). That is, the different analyses and parameters predicted greater magnitude of variability at different cycle instances. Therefore, timing of the variability peaks may provide additional information and should be investigated in the future. An explanation for the significantly lower variability during tube running compared to normal running after the intervention may be the constraint itself. If the constraints and the way they were used in this investigation somehow limited the freedom of movement, they may show the same effect as do overuse injuries as research on different diseases revealed reduced coupling variability. Although it is difficult to draw a direct comparison between our results and those of other studies (due to different couplings, examined populations or missing intervention), an adopted preventive mechanism due to the overpowering constraints would be in line with Hamill et al. (1999) or Heiderscheit et al. (2002), who studied patients with patella-femoral pain syndrome, and showed reduced joint coupling Please cite this article in press as: Haudum, A., et al. The effect of external perturbations on variability in joint coupling and single joint variability. Human Movement Science (2014), http://dx.doi.org/10.1016/ j.humov.2014.02.004
10
A. Haudum et al. / Human Movement Science xxx (2014) xxx–xxx 60
(a)
60
40 30 20 10 0 40
60
80
(c)
angular deviation (°)
angular deviation (°)
40
60
80
100
80
100
80
100
80
100
Percent of stride
(d)
50
30 20 10
40 30 20 10 0
20
40
60
80
20
100
Percent of stride
(e)
60
40
60
Percent of stride
(f)
50 angular deviation (°)
50 angular deviation (°)
10
60
0
40 30 20 10
40 30 20 10 0
0 20
40
60
80
20
100
Percent of stride
(g)
60
40
60
Percent of stride
(h)
50 angular deviation (°)
50 angular deviation (°)
20
20
Percent of stride
40
60
30
100
50
60
40
0 20
60
(b)
50 angular deviation (°)
angular deviation (°)
50
40 30 20 10
40 30 20 10 0
0 20
40
60
Percent of stride
80
100
20
40
60
Percent of stride
Fig. 4. Average CRPV plots for hip–knee and knee–ankle on a point-to point basis normalized to 100% of stride. Graphs a–d display hip–knee data, graphs e–h show knee–ankle data. The solid red thick line and the red thin solid line represent average and standard deviation of the tube-runs; the thick dashed black line and thin dashed black line represent average and standard deviation of the normal-runs, respectively. Graphs a and e display the first pre-test running block (i.e., min 0–2), graphs b and f the last pre-test running block (i.e., min 28–30), graphs and g the first post-test running block (i.e., min 0–2), and graphs d and h the last post-test running block (i.e., min 28–30). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
variability in the injured leg. Our post-test results are in also line with Ferber, Kendall, and Farr (2011), who investigated runners with patella-femoral pain syndrome and found reduced variability in the knee joint. They suggested that this reduced variability is the consequence of a kind of restoration Please cite this article in press as: Haudum, A., et al. The effect of external perturbations on variability in joint coupling and single joint variability. Human Movement Science (2014), http://dx.doi.org/10.1016/ j.humov.2014.02.004
A. Haudum et al. / Human Movement Science xxx (2014) xxx–xxx
11
of a more consistent movement pattern to reduce knee pain. In our study, the constraint in the form of the tubes would then be the equivalent of the injury, that is repetitive running with the tubes would then result in reduced variability (Haudum et al., 2011a). Another similarity, which was not investigated in our experiment but which was already demonstrated previously, supports this preventive mechanism. As Haudum et al. (2012), Haudum et al. (2012a), (2012b)showed in their tube running experiments, muscle activity was increased during tube running, which is also found in injured runners (Ferber et al., 2011). Another explanation for the missing effects (i.e., the fast adaptation within one test situation) in the test runs may be the characteristics of the tubes that were not altered throughout the test runs. The constraints were in a kind of way maintained unchanged (i.e., tube length or its position). Hence, we may speculate that the variability-increasing stimuli in the 30 min test run after the intervention were too weak (due to the learn-relevant adaptation following the tube running intervention) and diminished throughout the run. This would have resulted in an analogous tube running and control situation (Haudum et al., 2012b). This constant space may explain the similar variability in the constrained situation and the control situation after the intervention. A limitation of the study was the rather small sample size. Nevertheless, despite the small sample size, large effect sizes were demonstrated for both variability parameters. For future investigations, kinetics may provide details on the influences of the tubes at specific time points in the stride cycle.
References Batschelet, E. (1981). Circular statistics in biology. London; New York: Academic Press. Bernstein, N. A. (1967). The co-ordination and regulation of movements. Oxford: Pergamon Press. Birklbauer, J., Haudum, A., & Müller, E. (2006). Contextual-Interference-Approach versus Differential-Learning-Approach. In H. Hoppeler, T. Reilly, E. Tsolakidis, L. Gfeller, & S. Klossner (Eds.), 11th Annual Congress of the European College of Sport Science (p. 433). Lausanne, Switzerland: Sportverlag, Strauß, Cologne. Dierks, T. A., & Davis, I. (2007). Discrete and continuous joint coupling relationships in uninjured recreational runners. Clinical Biomechanics (Bristol, Avon), 22, 581–591. Fellin, R. E., Rose, W. C., Royer, T. D., & Davis, I. S. (2010). Comparison of methods for kinematic identification of footstrike and toe-off during overground and treadmill running. Journal of Science and Medicine in Sport, 13, 646–650. Ferber, R., Kendall, K. D., & Farr, L. (2011). Changes in knee biomechanics after a hip-abductor strengthening protocol for runners with patellofemoral pain syndrome. Journal of Athletic Training, 46, 142–149. Gittoes, M. J., & Wilson, C. (2010). Intralimb joint coordination patterns of the lower extremity in maximal velocity phase sprint running. Journal of Applied Biomechanics, 26, 188–195. Hamill, J., Haddad, J. M., & McDermott, W. J. (2000). Issues in quantifying variability from a dynamical systems perspective. Journal of Applied Biomechanics, 16, 407–418. Hamill, J., Palmer, C., & Van Emmerik, R. E. (2012). Coordinative variability and overuse injury. Sports, Medicine, Arthroscopy, Rehabilitation, Therapy and Technology, 4, 45. Hamill, J., van Emmerik, R. E. A., Heiderscheit, B. C., & Li, L. (1999). A dynamical systems approach to lower extremity running injuries. Clinical Biomechanics, 14, 297–308. Haudum, A., Birklbauer, J., Kröll, J., & Müller, E. (2011). Motor learning of gross-motor skills under variable practice conditions. Education, Physical Training, Sport, 80, 22–28. Haudum, A., Birklbauer, J., & Müller, E. (2011b). Physiological responses in ski tourers walking with support of elastic cords. Gazzetta Medica Italiana, 171, 785–794. Haudum, A., Birklbauer, J., & Müller, E. (2012a). The effect of an acute bout of rubber tube running constraint on kinematics and muscle activity. Journal of Sports Science and Medicine, 11, 459–467. Haudum, A., Birklbauer, J., & Müller, E. (2012b). The influence of external perturbations on running kinematics and muscle activity before and after accommodation. Journal of Sports Science and Medicine, 11, 582–591. Haudum, A., Birklbauer, J., & Müller, E. (2011a). Investigating the effect of a variability constraint on running kinematics and EMG. In T. Cable & K. Georg (Eds.), 16th annual congress of the european college of sport science (ECSS) (pp. 64–65). Liverpool: European College of Sport Science. Haudum, A., Birklbauer, J., Kröll, J., & Müller, E. (2012). Constraint-led changes in internal variability in running. Journal of Sport Science and Medicine Online, 11. Heiderscheit, B. C., Hamill, J., & van Emmerik, R. E. A. (2002). Variability of stride characteristics and joint coordination among individuals with unilateral patellofemoral pain. Journal of Applied Biomechanics, 18, 110–121. Hein, T., Schmeltzpfenning, T., Krauss, I., Maiwald, C., Horstmann, T., & Grau, S. (2012). Using the variability of continuous relative phase as a measure to discriminate between healthy and injured runners. Human Movement Science, 31, 683–694. Lamoth, C. J., Daffertshofer, A., Huys, R., & Beek, P. J. (2009). Steady and transient coordination structures of walking and running. Human Movement Science, 28, 371–386. Li, L., van den Bogert, E. C. H., Caldwell, G. E., van Emmerik, R. E. A., & Hamill, J. (1999). Coordination patterns of walking and running at similar speed and stride frequency. Human Movement Science, 18, 67–85. Miller, R. H., Chang, R., Baird, J. L., Van Emmerik, R. E. A., & Hamill, J. (2010). Variability in kinematic coupling assessed by vector coding and continuous relative phase. Journal of Biomechanics, 43, 2554–2560.
Please cite this article in press as: Haudum, A., et al. The effect of external perturbations on variability in joint coupling and single joint variability. Human Movement Science (2014), http://dx.doi.org/10.1016/ j.humov.2014.02.004
12
A. Haudum et al. / Human Movement Science xxx (2014) xxx–xxx
Mullineaux, R. D. (2007). Research methods: Sample size and variability effects on statistical power. In C. Payton & R. Bartlett (Eds.), Biomechanical evaluation of movement in sport and exercise: The British Association of Sport and Exercise Sciences guide (pp. 153–175). Abingdon, Oxon; New York, NY: Routledge. Nigg, B. M. (2001). The role of impact forces and foot pronation: A new paradigm. Clinical Journal of Sport Medicine, 11, 2–9. Schöllhorn, W. I., Hegen, P., & Davids, K. (2012). The nonlinear nature of learning – A differential learning approach. The Open Sports Sciences Journal, 5, 100–112. Schöllhorn, W. I., Mayer-Kress, G., Newell, K. M., & Michelbrink, M. (2009). Time scales of adaptive behavior and motor learning in the presence of stochastic perturbations. Human Movement Science, 28, 319–333. Stergiou, N., Jensen, J. L., Bates, B. T., Scholten, S. D., & Tzetzis, G. (2001a). A dynamical systems investigation of lower extremity coordination during running over obstacles. Clinical Biomechanics, 16, 213–221. Stergiou, N., Scholten, S. D., Jensen, J. L., & Blanke, D. (2001b). Intralimb coordination following obstacle clearance during running: The effect of obstacle height. Gait and Posture, 13, 210–220. Stöggl, T., Haudum, A., Birklbauer, J., Murrer, M., & Müller, E. (2010). Short and long term adaptation of variability during walking using unstable (Mbt) shoes. Clinical Biomechanics (Bristol, Avon), 25, 816–822. Tepavac, D., & Field-Fote, E. C. (2001). Vector coding: A technique for quantification of intersegmental coupling in multicyclic behaviors. Journal of Applied Biomechanics, 17, 259–270. Van Emmerik, R. E. A., & Van Wegen, E. E. H. (2000). On variability and stability in human movement. Journal of Applied Biomechanics, 16, 394–406. Wheat, J. S., & Glazier, P. S. (2006). Measuring coordination and variability in coordination. In K. Davids, S. Bennett, & K. Newell (Eds.), Movement system variability (pp. 167–181). Champaign, IL: Human Kinetics. Woltring, H. J. (1986). A fortran package for generalized, cross-validatory spline smoothing and differentiation. Advances in Engineering Software and Workstations, 8, 104–113.
Please cite this article in press as: Haudum, A., et al. The effect of external perturbations on variability in joint coupling and single joint variability. Human Movement Science (2014), http://dx.doi.org/10.1016/ j.humov.2014.02.004
Contents lists available at ScienceDirect
Human Movement Science journal homepage: www.elsevier.com/locate/humov
The effect of external perturbations on variability in joint coupling and single joint variability Anita Haudum ⇑, Jürgen Birklbauer, Erich Müller Department of Sport Science and Kinesiology, University of Salzburg, Salzburg, Austria Doppler Laboratory ‘‘Biomechanics in Skiing’’, Rifer Schlossallee 49, 5400 Hallein/Rif, Salzburg, Austria
a r t i c l e
i n f o
Article history: Available online xxxx PsycINFO classification: 2330 2343 3720 Keywords: Single joint variability Coupling variability Perturbations Running
a b s t r a c t This paper explores the effect of goal-oriented external perturbations created by elastic tubes attached to the hip and ankles on lower limb joint variability and hip–knee and knee–ankle coordination variability during running. Kinematics of ten healthy male runners were analysed prior to and following a 7-week tube running intervention while running with and one without this constraint. The training intervention was based on variable training aspects to increase within-movement variability and adaptability of the running pattern. To analyse the effects of the tubes on the running pattern, the phase plot vector length deviation (i.e., the standard deviation of the phase plot vector length) for the within-joint variability and the continuous relative phase variability for the joint coupling variability were calculated. Results revealed acute increases of variability in both parameters. However, after the intervention, variability of the tube running situation returned to normal for all couplings and joints except the knee. No transfer effects to normal running were observed. This suggests very rapid adaptations to such perturbations. In the long-term, it may ask for more or different variations. Ó 2014 Elsevier B.V. All rights reserved.
⇑ Corresponding author. Address: Department of Sport Science and Kinesiology, University of Salzburg, Rifer Schlossallee 49, 5400 Hallein/Rif, Salzburg, Austria. Tel.: +43 664 7505 7177; fax: +43 662 60 44 614. E-mail address: [email protected] (A. Haudum). http://dx.doi.org/10.1016/j.humov.2014.02.004 0167-9457/Ó 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: Haudum, A., et al. The effect of external perturbations on variability in joint coupling and single joint variability. Human Movement Science (2014), http://dx.doi.org/10.1016/ j.humov.2014.02.004
2
A. Haudum et al. / Human Movement Science xxx (2014) xxx–xxx
1. Introduction Running has always been a very popular recreational activity. However, improved fitness and growing ambition lead to an increased amount and intensity of training that is often associated with overuse injuries in the hip, knee or ankle (Hamill, van Emmerik, Heiderscheit, & Li, 1999; Heiderscheit, Hamill, & van Emmerik, 2002; Hein et al., 2012; Stergiou, Jensen, Bates, Scholten, & Tzetzis, 2001). Due to pain-related movement, joint coordination variability has been shown to decrease as a kind of protective mechanism (Van Emmerik & Van Wegen, 2000). The loss of variability, however, results in reduced adaptability or flexibility, which are necessary for optimal functioning of healthy systems (Heiderscheit et al., 2002). As the mastering and organization of the many degrees of freedom allows the adoption of functional movement patterns (Bernstein, 1967), analysing changes due to continuous training in healthy individuals may reveal possible changes in joint coordination variability and help prevent injuries. Hence, a large number of running studies (e.g., Dierks & Davis, 2007; Gittoes & Wilson, 2010; Hamill et al., 1999; Heiderscheit et al., 2002; Hein et al., 2012) investigated either two populations (i.e., one healthy and one pathological) or report only single measurements done with respect to lower limb joint kinematics and coordination variability. However, limited research is available focusing on the application of stimuli intended to increase variability to regain or maintain a healthy amount of coordination or coupling variability. Since it is largely accepted that variability is a functional entity that helps establish optimal states of coordination (Hamill, Palmer, & Van Emmerik, 2012; Wheat & Glazier, 2006), investigations that apply perturbations to the movement pattern may help to understand whether inducing external goaldirected perturbations enables an increase of functional variability within the movement pattern (Haudum, Birklbauer, Kröll, & Müller, 2011, 2012). That is, through the application of additional variability, the movement pattern may again become more flexible and, therefore, more adaptable to external perturbations (Haudum et al., 2011, 2012; Schöllhorn, Hegen, & Davids, 2012; Schöllhorn, Mayer-Kress, Newell, & Michelbrink, 2009). This may further help to move more economically due to a transition to a new or at least slightly altered behavioral pattern (Haudum, Birklbauer, & Müller, 2011b). In running, where many degrees of freedom are orchestrated to a stable, coordinated and skillful movement, this approach may include the effect of running under changing conditions or responses to intervention programs in a healthy runners’ population (Stergiou, Jensen et al., 2001). An example for changing conditions or constraints can be obstacles (Stergiou, Jensen et al., 2001), type of shoes (Stöggl, Haudum, Birklbauer, Murrer, & Müller, 2010) or the application of elastic tubes (Haudum, Birklbauer, & Müller, 2012b; Haudum et al., 2011; Haudum et al., 2012), which may allow a more precise addition of variability. Several running studies are available on acute influences of varied conditions, where coordination has been examined such as obstacles in the track (Stergiou, Jensen et al., 2001; Stergiou, Scholten, Jensen, & Blanke, 2001) or elastic tubes applied to the lower extremities (e.g., Haudum et al., 2012). To our knowledge, only few intervention studies exist where joint coordination was examined especially under varying conditions (e.g., Haudum, Birklbauer, & Müller, 2011a; Haudum et al., 2012b). A limited number of studies, however, have employed intervention-induced variability to maintain or regain a functional level of variability in healthy individuals in joint coupling variability (Haudum et al., 2012b), which may provide relevant insights into possible alterations due to the prevailing constraints (Hein et al., 2012). In addition, the effect of variable constraints applied in acute situations or over a longer period of time may differ, since a new and possibly more appropriate movement pattern may be developed. In a first step and to clarify the role of a dynamic constraint in the form of elastic tubes on joint motion and joint coupling variability in a healthy population, we employed kinematic motion analyses to determine the effects of such an intervention. The elastic tubes were used to apply perturbations to the running pattern so that within movement variability is increased but the running patter is not rendered. The tubes are used to functionally achieve an optimal level of within-movement variability and to support the shaping or initiating of emergent movement patterns. Due to the sensitivity of the
Please cite this article in press as: Haudum, A., et al. The effect of external perturbations on variability in joint coupling and single joint variability. Human Movement Science (2014), http://dx.doi.org/10.1016/ j.humov.2014.02.004
A. Haudum et al. / Human Movement Science xxx (2014) xxx–xxx
3
initial conditions, the tubes have a different impact and, therefore, ask for permanent adaptations to their perturbations. Since the tubes demand a flexible movement pattern and permanent adaptations, we hypothesize that movement variability is influenced by the harness. Hence, the main objective of the study was to determine the influences of dynamic constraints (i.e., elastic tubes attached to the legs to slightly increase within-movement variability) prior to and following a 7-week training intervention on continuous relative phase variability (CRPV) and single joint variability (SJV) and whether there are differences in adaptation between CRPV and SJV. Since coordination is the functional link between muscles, joints and body segments to achieve a desired movement outcome (Tepavac & Field-Fote, 2001), the investigation of both may be of high importance. Since the magnitude of variability depends on the calculations being used, it need not be the case that all parameters reflect the same direction (Mullineaux, 2007). Further, as the results may differ when analysing the entire stride or when analysing stance or swing separately (Hamill et al. 1999), stride as well as stance and swing were analysed. We hypothesized that tube running will result in higher acute variability than normal running and that the initially higher tube running variability will decrease throughout the intervention due to tube running adaptation. We further hypothesize a transfer effect of the tube running intervention on normal running variability due to a generalization of the tube running pattern because of the proximity of the two running conditions (i.e., running with the tubes and running without tubes). With respect to the amount of influences of the variable constraint, we expected differences between entire stride, stance phase or swing calculations. 2. Methods 2.1. Participants Thirteen recreational runners gave their written informed consent to participate in the study. All runners had either participated in prior treadmill running interventions or were used to train on a treadmill and therefore considered as treadmill experienced. Of the original sample size, ten runners (mean age: 26.1 ± 7.1 years, mean height: 177.1 ± 7.2 cm, mean weight: 72.0 ± 6.6 kg) could be included in the analysis due to illness and injury of the 3 other participants. None of the runners had prior experience in running with the constraints (Fig. 1). The study was approved by the local ethics committee. 2.2. Training device A padded harness (Fig. 1; Tendybelt, Salzburg, Austria) was used to fix the tubes (Thera-BandÒ GmbH, Dornburg-Frickhofen, Germany) at both ankles (i.e., at the heel tab of the running shoes) and the lower back (ilio-sacral joint). Due to the results observed in pre-investigations, the tube length was set at 40% of the individual leg length (Haudum et al., 2012; Haudum, Birklbauer, & Müller, 2012a; Haudum et al., 2012b). The tubes are attached in order to apply variability to the running pattern, whereby the tubes’ resistance should in no way render running with the current running pattern impossible; rather, the tubes’ resistance should create additional variability within the movement skill. The functional role of the tubes is to achieve an optimal level of within-movement variability and to support the shaping or initiating of emergent movement patterns. The chosen tube length should ensure that small perturbations would increase within running variability and require permanent adaptations to the tube perturbations. 2.3. Training intervention Participants completed 18 training sessions of tube running on a treadmill (10.5 km.h 1 and 0% grade) over 7 weeks. Alternately, 2 and 3 sessions were run each week. The session duration was increased from 45 min for sessions 1–2 to 50 min for sessions 3–6 and to 55 min for sessions 7–18. The exercises during the training session were the same for each participant. Please cite this article in press as: Haudum, A., et al. The effect of external perturbations on variability in joint coupling and single joint variability. Human Movement Science (2014), http://dx.doi.org/10.1016/ j.humov.2014.02.004
4
A. Haudum et al. / Human Movement Science xxx (2014) xxx–xxx
Fig. 1. One participant running with the training device.
2.3.1. Training contents The aim of the training contents was to best possibly apply variability in the form of external perturbations within the movement pattern. The versatile exercises should help establish the individually most effective and optimal movement pattern. The intervention included different tube running exercises and was conducted according to differential learning (Schöllhorn et al., 2009) and training guidelines proposed by Birklbauer, Haudum, and Müller (2006). The differential learning approach (Schöllhorn et al., 2009) is considering movement variations as the basis of learning rather than movement repetitions and it is, therefore, taking advantage of the necessity of fluctuations or errors for learning (Schöllhorn, Hegen & Davids, 2012). These fluctuations through the versatile exercises represent movement differences that require constant reactions and adaptations to the varying constraints and which according to Schöllhorn provided essential information for refining or a self-organization of a better movement pattern. The guidelines (Birklbauer et al., 2006) were used to structure the exercises with respect to the amount of variability and where variability was induced. These guidelines emphasise on key points Please cite this article in press as: Haudum, A., et al. The effect of external perturbations on variability in joint coupling and single joint variability. Human Movement Science (2014), http://dx.doi.org/10.1016/ j.humov.2014.02.004
A. Haudum et al. / Human Movement Science xxx (2014) xxx–xxx
5
of the movement to be considered when applying variability and structure exercises (Haudum et al., 2012b). Examples of these guidelines were that the induced variability must be applied so that it supports the development of the movement, which means that the variability is intervention-induced applied; then the range of difference should decrease throughout the learning/intervention process. That is, a beginner needs larger differences between two consecutive exercises to perceive the difference between the two stimuli while an advanced runner or expert with the tubes can perceive even smaller differences between exercises due to his advanced perception of tube stimuli (i.e., larger differences at the beginning and smaller ones at the end of learning, which actually results in the same perceived differences) (Haudum et al., 2012b). This should help the runners utilize the tubes (i.e., the effect of reactive phenomena). The various exercises were chosen with respect to three key classes to provoke perturbations within the movement pattern. Each one of the three classes had three sub-aspects which were: (A) Intended variations with the participants focusing on the tubes: active work against the tubes; optimal work (i.e., use the tubes forces); supporting the tubes reactions (examples for such instructions were: running without completely stretching the tubes in any situation or bring the knee more upward and forward than usual); (B) Differences between the left and right leg in the position of the tubes: the tubes being attached between hip and shank; the tubes being attached between hip and the laces of the running shoe; no tubes attached; (C) Differences in the length of the tubes: shorter tubes so that running for a given time is at the utmost possible; the tube position as in the test runs (between hip and ankle); running without tubes. 2.4. Procedure Pre- and post-tests were held prior to and after a 7-week training intervention. Both times, 2 30 min tests (1 with and 1 without tubes) were run on a motorized treadmill (HP Cosmos Quasar 170/65, Traunstein, Germany). Both experimental running conditions were performed at 10.5 km.h 1 at 0% grade. This speed was chosen due to preliminary experiments (Haudum et al., 2012; Haudum et al., 2012a, 2012b) on the tubes behavior at different speeds and due to the participants’ endurance performance. Prior to the start of each test, runners were instructed to run without tubes for 5 min to warm up (8.5 km.h 1; 0% grade). Five runners started with the tubes, the other 5 with normal running. In the normal running situation, runners wore the harness but without the tubes. Runners were provided a 60-min recovery time between the two tests. Kinematics was recorded in all runs. 2.5. Data collection and processing Kinematic data were sampled in 2-min blocks starting at minutes 0, 3, 13, 16, 25 and 28. Of each 2min block the first 90 strides of recorded data were selected for further analysis, except in the first 2min block of each test run (i.e., min 0–2). Here, the first 10 strides were not included; however, the subsequent strides (i.e., 11–100) were selected for analysis to ensure that treadmill acceleration was finished and no influences due to different speed distorted the data. To collect kinematic data, an 8-camera Vicon three-dimensional motion analysis system (Vicon Peak, Oxford, UK) was used to analyse ankle, knee and hip movement in the sagittal plane. A total of 41 markers with a diameter of 14 mm were affixed to specific anatomical landmarks according to the Plug-In-Gait model (Plug-In-Gait Marker Set, Vicon Peak, Oxford, UK) (Haudum et al., 2012a, 2012b). The 3D-trajectories of all markers were analyzed utilizing Nexus software (Nexus 1.3, Vicon, Oxford, UK). Data were sampled at a rate of 250 Hz. The manually labelled marker trajectories were smoothed via a Woltring routine (MSE value of 10) (Woltring, 1986) and kinematics were calculated using the Plug-in-Gait model (Haudum et al., 2012b). Joint angles for the right leg ankle (plantar-dorsiflexion), knee (flexion–extension), and hip (flexion–extension) and angular velocities were exported to IKE-master (IKE-Software Solutions, Salzburg, Austria) for further analyses. Each stride cycle was defined from right heel contact to the next right heel contact. In addition, stance phase and swing were separately analysed. Heel contact and toe off were identified using the vertical velocities and the position profiles of the heel and toe markers as described by Fellin, Rose, Royer, and Davis (2010) and Lamoth, Daffertshofer, Huys, and Beek (2009), respectively. All angular Please cite this article in press as: Haudum, A., et al. The effect of external perturbations on variability in joint coupling and single joint variability. Human Movement Science (2014), http://dx.doi.org/10.1016/ j.humov.2014.02.004
6
A. Haudum et al. / Human Movement Science xxx (2014) xxx–xxx
data were interpolated to 101 data points and normalized to 0–100% of stride, stance phase or swing (Haudum et al., 2012b). 2.6. Data analysis Both intralimb coupling variability and single joint variability analyses were performed to determine movement variability of the right lower extremity. 2.6.1. Joint coupling variability The intralimb inter-joint coupling variability was assessed using continuous relative phase variability (CRPV). The coordination of the hip and knee joint and the knee and ankle joint was analysed according to the method of Hamill et al. (1999). That is, phase plots of the calculated joint angles were calculated with each phase plot consisting of the angle on the horizontal axis and the respective angular velocity on the vertical axis. In order to adjust the amplitude differences in the different joint angle (i.e., hip, knee, ankle joint) in the ranges of motion and centre the plots about the origin, angular displacement normalized to a range of 1 to +1. The angular velocity was normalized to the absolute maximum value to maintain zero velocity at the origin (Hamill, Haddad, & McDermott, 2000; Li, van den Bogert, Caldwell, van Emmerik, & Hamill, 1999). The normalization procedure was always conducted with respect to the maximum value over all strides within each 2-min block to maintain the true spatial properties. For all three joints, phase plots were employed to compare the two running situations with the phase angle being defined as the angle between the right horizontal and a line drawn from the origin to a specific data point. Subtracting the distal joint from the proximal, the hip knee and knee ankle continuous relative phases were calculated in the range of 0–360° avoiding discontinuities (Hamill et al., 2000; Wheat & Glazier, 2006). To determine CRPV, the angular deviation, which is the circular equivalent to the standard deviation in linear statistics, was calculated due to the directional nature of continuous relative phase (Batschelet, 1981; Miller, Chang, Baird, Van Emmerik, & Hamill, 2010). For each participant, the calculated individual CRPV was averaged across the complete stride, stance phase or swing for statistical analyses. 2.6.2. Single joint variability To separately analyse variability of each investigated joint, the phase plot vector length deviation (i.e., the dispersion of the vector lengths in the normalized phase plots) was applied. The variability for these linear data was calculated on a point-by-point basis across all 90 strides for the hip, knee and ankle joints in the form of the standard deviation. Similar to CRPV, the single joint variability (SJV) was averaged across the complete stride, stance phase or swing for statistical analyses. 2.7. Statistical analysis The IBM SPSS software Ver. 20.0. (SPSS Inc., Chicago, IL) was used for statistical analyses. For each participant, the mean joint coupling variability and the mean single joint variability was calculated in each running condition and for each 2-min block at each test (pre- and post-test). We performed 2 2 6 repeated measures ANOVAs (RMANOVA) with situation (tubes or normal running), time (pre- and post-test) and data blocks (2 min data block) as factors for both CRPV and SJV. 2 6 RMANOVA were calculated to estimate differences between the different pre- and post-test runs. Significant differences were reported at p < .05. Effect size partial eta squared (pg2) was additionally calculated. 3. Results 3.1. Continuous relative phase variability Looking at the group means of the averaged CRPV, which are graphically presented for stride (Fig. 2a and b), stance phase (Fig. 2c and d) and swing (Fig. 2e and f), the significant increases in Please cite this article in press as: Haudum, A., et al. The effect of external perturbations on variability in joint coupling and single joint variability. Human Movement Science (2014), http://dx.doi.org/10.1016/ j.humov.2014.02.004
A. Haudum et al. / Human Movement Science xxx (2014) xxx–xxx
7
Fig. 2. Average CRPV graphs for hip–knee (left column) and knee–ankle (right column) ±95% confidence interval. Graphs a and b represent averages for stride, graphs c and d averages for stance phase, and graphs e and f averages for swing. s = normal running pre-test; d = tube running pre-test; = normal running post-test; = tube running post-test.
CRP variability in the tube run ranged between 120% and 130% compared to the normal running level at the pre-test (p < .05; pg2 > .42). The CRPV was rather similar in both running conditions toward the end of the pre-test blocks. Separate analyses unveiled that the stride and stance phase data showed an increased variability for the hip–knee coupling in the tube running test at the pre-test which was reversed at the post test runs in comparison to normal running (p < .05; pg2 > .73). That is, the post-test unveiled approximately 15% less variability in the tube situation. While there was no significant result in the pre-test for the swing data, there was a considerable difference for the hip–knee CRPV in the post-test runs reasoned from the reduced tube running CRPV, which marginally failed significance (p < .06; pg2 > .55). For the knee–ankle complex, swing and stride data showed an initial increase to 120% of the normal running variability in the pre-test tube situation but apart from the initial increase data were Please cite this article in press as: Haudum, A., et al. The effect of external perturbations on variability in joint coupling and single joint variability. Human Movement Science (2014), http://dx.doi.org/10.1016/ j.humov.2014.02.004
8
A. Haudum et al. / Human Movement Science xxx (2014) xxx–xxx
rather similar for both tests which resulted in no significant differences (p > .05; pg2 < .15). The ankle– knee coupling variability during stance phase was reduced in both post-test runs compared to the pretest runs (p < .05; pg2 > .66). No further significant differences were found for this coupling.
3.2. Single joint variability Results for the SJV are displayed in Fig. 3 for stride (3a–c), stance phase (3d–f) and swing (3g–i). The variability increases due to the tubes were up to 170% for the knee compared to the normal running variability (p < .05; pg2 > .54) and for the hip (p < .05; pg2 > .56) and ankle up to 140% compared to normal running variability. The intervention-induced changes were very similar for all three, stride, stance phase and swing. That is, for both stance phase and swing as well as the total stride data, the increased variability in all three joints in the tube running pre-test was annihilated in the post-test runs (p < .05; pg2 > .45). For the knee joint, variability for normal running was significantly higher at post-test compared to the variability during tube running (p < .05; pg2 > .50). For both, SJV and CRPV, the significant 2 2 6 interaction is based on the initial increases in the tube running situation (p < .05; pg2 > .32). While the increases over several 2-min blocks in the pretest tube run, in the post-test tube run the initial 2-min blocks showed increased variability, which returned to normal rather fast. While the tube running variability changed from pre- to post-test, no transfer effects to normal running were found for either CRPV or SJV (p > .05; pg2 < .10).
Fig. 3. Average SJV graphs for hip, knee and ankle ±95% confidence interval. Graph a–c represent averages for stride, graphs d–f averages for stance phase, and g–h averages for swing. s = normal running pre-test; d = tube running pre-test; = normal running post-test; = tube running post-test.
Please cite this article in press as: Haudum, A., et al. The effect of external perturbations on variability in joint coupling and single joint variability. Human Movement Science (2014), http://dx.doi.org/10.1016/ j.humov.2014.02.004
A. Haudum et al. / Human Movement Science xxx (2014) xxx–xxx
9
4. Discussion The aim of this study was to investigate the effect of a varied training intervention on lower limb coordination variability (CRPV) and single joint variability (SJV) in healthy male runners. Using continuous relative phase and single joint analyses, we examined variability from two different perspectives. Based on our results and with the caution needed to compare different analyses techniques due to their different quantitative findings or requirements (Mullineaux, 2007, p. 107), an influence of the variability constraint was evident in almost all evaluated parameters, especially for the pre-test results. The magnitude of variability and of the constraints’ influence, however, differed between within-joint analyses and between-joint analyses. That is, the single joint approach exhibited a higher vulnerability (or stronger influence due) to the constraints than analyses between joints. It can be assumed that the within-joint variability represents a more flexible movement level while the betweenjoint parameters indicate that a robust movement path does exist (Haudum et al., 2012a, 2012b; Nigg, 2001). With respect to co-variation, which reflects the coupling between two joints, it can be stated that the movement pattern for the analysed skill seem well defined. Nevertheless, this difference between observation levels supports the selection of both analysis strategies as the amount of the actual influence varied between the methods. The tendencies of our data are in line with data from another locomotion intervention (i.e., walking), where another kind of variability constraint was used. In this intervention study on walking with unstable MBT shoes (Stöggl et al., 2010), the authors showed that after several weeks of walking with MBT shoes, which also intends to increase variability, repeated walking with MBT shoes also led to reduced variability during MBT walking. Both studies support the perfect adaption ability of the human system but they also demonstrate that it requires also new stimuli to keep the level of variability higher. Since the data showed slightly increased variability at the initial data block of the post-test tube run, different tube applications during the test runs might have revealed different effects on the SJV or CRPV. The increased variability during the first exposure supports results of other studies (Haudum et al., 2012; Haudum et al., 2012a). One explanation for the largely increased SJV and CRPV in the pre-tube runs is that the higher variability may serve an attempt to adjust the running pattern in the variability device-condition (Heiderscheit et al., 2002). Due to the observed decrease of variability especially in the SJV parameters, this strategy might be successful in the beginning as a kind of exploratory behavior to accommodate to the tubes (Haudum et al., 2012; Haudum et al., 2011a). However, while the strategy may be initially effective, tube running practice resulted in similar variability of both tube running and normal running after the intervention. Since normal running was not influenced by the intervention, a normalization of the tube running pattern occurred because the variability in the tube situation rather fast approached the variability level of the normal running situation (i.e., similar variability in normal running and tube running kinematics) and the assumption that such intervention leads to behavioral changes in the form of functional CRPV or SJV increases could not be confirmed. When examining CRPV averaged over the entire stride, stance phase or swing, results did not show significant differences, though tendencies towards different amounts of variability. The graphical illustrations (Fig. 4) demonstrated that despite the missing statistically significant differences between the two situations, the temporal structure or the variability peaks differed (Hein et al., 2012; Mullineaux, 2007). That is, the different analyses and parameters predicted greater magnitude of variability at different cycle instances. Therefore, timing of the variability peaks may provide additional information and should be investigated in the future. An explanation for the significantly lower variability during tube running compared to normal running after the intervention may be the constraint itself. If the constraints and the way they were used in this investigation somehow limited the freedom of movement, they may show the same effect as do overuse injuries as research on different diseases revealed reduced coupling variability. Although it is difficult to draw a direct comparison between our results and those of other studies (due to different couplings, examined populations or missing intervention), an adopted preventive mechanism due to the overpowering constraints would be in line with Hamill et al. (1999) or Heiderscheit et al. (2002), who studied patients with patella-femoral pain syndrome, and showed reduced joint coupling Please cite this article in press as: Haudum, A., et al. The effect of external perturbations on variability in joint coupling and single joint variability. Human Movement Science (2014), http://dx.doi.org/10.1016/ j.humov.2014.02.004
10
A. Haudum et al. / Human Movement Science xxx (2014) xxx–xxx 60
(a)
60
40 30 20 10 0 40
60
80
(c)
angular deviation (°)
angular deviation (°)
40
60
80
100
80
100
80
100
80
100
Percent of stride
(d)
50
30 20 10
40 30 20 10 0
20
40
60
80
20
100
Percent of stride
(e)
60
40
60
Percent of stride
(f)
50 angular deviation (°)
50 angular deviation (°)
10
60
0
40 30 20 10
40 30 20 10 0
0 20
40
60
80
20
100
Percent of stride
(g)
60
40
60
Percent of stride
(h)
50 angular deviation (°)
50 angular deviation (°)
20
20
Percent of stride
40
60
30
100
50
60
40
0 20
60
(b)
50 angular deviation (°)
angular deviation (°)
50
40 30 20 10
40 30 20 10 0
0 20
40
60
Percent of stride
80
100
20
40
60
Percent of stride
Fig. 4. Average CRPV plots for hip–knee and knee–ankle on a point-to point basis normalized to 100% of stride. Graphs a–d display hip–knee data, graphs e–h show knee–ankle data. The solid red thick line and the red thin solid line represent average and standard deviation of the tube-runs; the thick dashed black line and thin dashed black line represent average and standard deviation of the normal-runs, respectively. Graphs a and e display the first pre-test running block (i.e., min 0–2), graphs b and f the last pre-test running block (i.e., min 28–30), graphs and g the first post-test running block (i.e., min 0–2), and graphs d and h the last post-test running block (i.e., min 28–30). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
variability in the injured leg. Our post-test results are in also line with Ferber, Kendall, and Farr (2011), who investigated runners with patella-femoral pain syndrome and found reduced variability in the knee joint. They suggested that this reduced variability is the consequence of a kind of restoration Please cite this article in press as: Haudum, A., et al. The effect of external perturbations on variability in joint coupling and single joint variability. Human Movement Science (2014), http://dx.doi.org/10.1016/ j.humov.2014.02.004
A. Haudum et al. / Human Movement Science xxx (2014) xxx–xxx
11
of a more consistent movement pattern to reduce knee pain. In our study, the constraint in the form of the tubes would then be the equivalent of the injury, that is repetitive running with the tubes would then result in reduced variability (Haudum et al., 2011a). Another similarity, which was not investigated in our experiment but which was already demonstrated previously, supports this preventive mechanism. As Haudum et al. (2012), Haudum et al. (2012a), (2012b)showed in their tube running experiments, muscle activity was increased during tube running, which is also found in injured runners (Ferber et al., 2011). Another explanation for the missing effects (i.e., the fast adaptation within one test situation) in the test runs may be the characteristics of the tubes that were not altered throughout the test runs. The constraints were in a kind of way maintained unchanged (i.e., tube length or its position). Hence, we may speculate that the variability-increasing stimuli in the 30 min test run after the intervention were too weak (due to the learn-relevant adaptation following the tube running intervention) and diminished throughout the run. This would have resulted in an analogous tube running and control situation (Haudum et al., 2012b). This constant space may explain the similar variability in the constrained situation and the control situation after the intervention. A limitation of the study was the rather small sample size. Nevertheless, despite the small sample size, large effect sizes were demonstrated for both variability parameters. For future investigations, kinetics may provide details on the influences of the tubes at specific time points in the stride cycle.
References Batschelet, E. (1981). Circular statistics in biology. London; New York: Academic Press. Bernstein, N. A. (1967). The co-ordination and regulation of movements. Oxford: Pergamon Press. Birklbauer, J., Haudum, A., & Müller, E. (2006). Contextual-Interference-Approach versus Differential-Learning-Approach. In H. Hoppeler, T. Reilly, E. Tsolakidis, L. Gfeller, & S. Klossner (Eds.), 11th Annual Congress of the European College of Sport Science (p. 433). Lausanne, Switzerland: Sportverlag, Strauß, Cologne. Dierks, T. A., & Davis, I. (2007). Discrete and continuous joint coupling relationships in uninjured recreational runners. Clinical Biomechanics (Bristol, Avon), 22, 581–591. Fellin, R. E., Rose, W. C., Royer, T. D., & Davis, I. S. (2010). Comparison of methods for kinematic identification of footstrike and toe-off during overground and treadmill running. Journal of Science and Medicine in Sport, 13, 646–650. Ferber, R., Kendall, K. D., & Farr, L. (2011). Changes in knee biomechanics after a hip-abductor strengthening protocol for runners with patellofemoral pain syndrome. Journal of Athletic Training, 46, 142–149. Gittoes, M. J., & Wilson, C. (2010). Intralimb joint coordination patterns of the lower extremity in maximal velocity phase sprint running. Journal of Applied Biomechanics, 26, 188–195. Hamill, J., Haddad, J. M., & McDermott, W. J. (2000). Issues in quantifying variability from a dynamical systems perspective. Journal of Applied Biomechanics, 16, 407–418. Hamill, J., Palmer, C., & Van Emmerik, R. E. (2012). Coordinative variability and overuse injury. Sports, Medicine, Arthroscopy, Rehabilitation, Therapy and Technology, 4, 45. Hamill, J., van Emmerik, R. E. A., Heiderscheit, B. C., & Li, L. (1999). A dynamical systems approach to lower extremity running injuries. Clinical Biomechanics, 14, 297–308. Haudum, A., Birklbauer, J., Kröll, J., & Müller, E. (2011). Motor learning of gross-motor skills under variable practice conditions. Education, Physical Training, Sport, 80, 22–28. Haudum, A., Birklbauer, J., & Müller, E. (2011b). Physiological responses in ski tourers walking with support of elastic cords. Gazzetta Medica Italiana, 171, 785–794. Haudum, A., Birklbauer, J., & Müller, E. (2012a). The effect of an acute bout of rubber tube running constraint on kinematics and muscle activity. Journal of Sports Science and Medicine, 11, 459–467. Haudum, A., Birklbauer, J., & Müller, E. (2012b). The influence of external perturbations on running kinematics and muscle activity before and after accommodation. Journal of Sports Science and Medicine, 11, 582–591. Haudum, A., Birklbauer, J., & Müller, E. (2011a). Investigating the effect of a variability constraint on running kinematics and EMG. In T. Cable & K. Georg (Eds.), 16th annual congress of the european college of sport science (ECSS) (pp. 64–65). Liverpool: European College of Sport Science. Haudum, A., Birklbauer, J., Kröll, J., & Müller, E. (2012). Constraint-led changes in internal variability in running. Journal of Sport Science and Medicine Online, 11. Heiderscheit, B. C., Hamill, J., & van Emmerik, R. E. A. (2002). Variability of stride characteristics and joint coordination among individuals with unilateral patellofemoral pain. Journal of Applied Biomechanics, 18, 110–121. Hein, T., Schmeltzpfenning, T., Krauss, I., Maiwald, C., Horstmann, T., & Grau, S. (2012). Using the variability of continuous relative phase as a measure to discriminate between healthy and injured runners. Human Movement Science, 31, 683–694. Lamoth, C. J., Daffertshofer, A., Huys, R., & Beek, P. J. (2009). Steady and transient coordination structures of walking and running. Human Movement Science, 28, 371–386. Li, L., van den Bogert, E. C. H., Caldwell, G. E., van Emmerik, R. E. A., & Hamill, J. (1999). Coordination patterns of walking and running at similar speed and stride frequency. Human Movement Science, 18, 67–85. Miller, R. H., Chang, R., Baird, J. L., Van Emmerik, R. E. A., & Hamill, J. (2010). Variability in kinematic coupling assessed by vector coding and continuous relative phase. Journal of Biomechanics, 43, 2554–2560.
Please cite this article in press as: Haudum, A., et al. The effect of external perturbations on variability in joint coupling and single joint variability. Human Movement Science (2014), http://dx.doi.org/10.1016/ j.humov.2014.02.004
12
A. Haudum et al. / Human Movement Science xxx (2014) xxx–xxx
Mullineaux, R. D. (2007). Research methods: Sample size and variability effects on statistical power. In C. Payton & R. Bartlett (Eds.), Biomechanical evaluation of movement in sport and exercise: The British Association of Sport and Exercise Sciences guide (pp. 153–175). Abingdon, Oxon; New York, NY: Routledge. Nigg, B. M. (2001). The role of impact forces and foot pronation: A new paradigm. Clinical Journal of Sport Medicine, 11, 2–9. Schöllhorn, W. I., Hegen, P., & Davids, K. (2012). The nonlinear nature of learning – A differential learning approach. The Open Sports Sciences Journal, 5, 100–112. Schöllhorn, W. I., Mayer-Kress, G., Newell, K. M., & Michelbrink, M. (2009). Time scales of adaptive behavior and motor learning in the presence of stochastic perturbations. Human Movement Science, 28, 319–333. Stergiou, N., Jensen, J. L., Bates, B. T., Scholten, S. D., & Tzetzis, G. (2001a). A dynamical systems investigation of lower extremity coordination during running over obstacles. Clinical Biomechanics, 16, 213–221. Stergiou, N., Scholten, S. D., Jensen, J. L., & Blanke, D. (2001b). Intralimb coordination following obstacle clearance during running: The effect of obstacle height. Gait and Posture, 13, 210–220. Stöggl, T., Haudum, A., Birklbauer, J., Murrer, M., & Müller, E. (2010). Short and long term adaptation of variability during walking using unstable (Mbt) shoes. Clinical Biomechanics (Bristol, Avon), 25, 816–822. Tepavac, D., & Field-Fote, E. C. (2001). Vector coding: A technique for quantification of intersegmental coupling in multicyclic behaviors. Journal of Applied Biomechanics, 17, 259–270. Van Emmerik, R. E. A., & Van Wegen, E. E. H. (2000). On variability and stability in human movement. Journal of Applied Biomechanics, 16, 394–406. Wheat, J. S., & Glazier, P. S. (2006). Measuring coordination and variability in coordination. In K. Davids, S. Bennett, & K. Newell (Eds.), Movement system variability (pp. 167–181). Champaign, IL: Human Kinetics. Woltring, H. J. (1986). A fortran package for generalized, cross-validatory spline smoothing and differentiation. Advances in Engineering Software and Workstations, 8, 104–113.
Please cite this article in press as: Haudum, A., et al. The effect of external perturbations on variability in joint coupling and single joint variability. Human Movement Science (2014), http://dx.doi.org/10.1016/ j.humov.2014.02.004