- Email: [email protected]
Effects of head extension on undisturbed upright stance control in humans
Gait and Posture 21 (2005) 318–325
Effects of head extension on undisturbed upright stance control in humans Nicolas Vuillerme∗ , Patrice Rougier Laboratoire de Mod´elisation des Activit´es Sportives, Universit´e de Savoie, Domaine Universitaire, 73376 Le Bourget du Lac cedex, France Received 17 November 2003; accepted 1 April 2004
Abstract The purpose of the present experiment was to investigate whether and how the head extended posture, commonly encountered in many routine activities, affects undisturbed upright stance control mechanisms in humans. Sixteen young healthy adults stood feet together, with their eyes closed and were asked to sway as little as possible in two Neutral and Extended head conditions. Centre of pressure (CP) displacements, recorded using a force platform, were used to compute the motions of the vertical projection of the centre of gravity (CGv ) and those of the difference CP − CGv . A time-domain analysis shows increased mean velocity and surface covered by the trajectory of both elementary motions in the Extended head condition. A frequency analysis also reveals increased root mean squares on the CP − CGv motions, suggesting increased muscular activity in the Extended head condition. Furthermore, similar changes occur on CGv motions. Finally, modelling these trajectories as a fractional Brownian motion process demonstrates increased spatial transition point co-ordinates at which the corrective process is initiated and a more deterministic control mechanism in this corrective process involving CGv motions in the Extended head position. Together, the present findings suggest that head extension position yields a reorganisation of the control mechanisms for maintaining undisturbed upright stance. © 2004 Elsevier B.V. All rights reserved. Keywords: Undisturbed upright stance control; Head position; Centre of pressure; Centre of gravity; Frequency analysis; Fractional Brownian motion modelling
1. Introduction The ability to maintain an upright stance is essential in gait and in the initiation of voluntary movements that are vital not only for professional, sportive but also for daily living activities. As we move in a wide range of situations, under changing tasks and environmental demands, postural control entails adapting to various task parameters (e.g., [1]). However, imbalance might occur under specific situations, conditions and/or postures. For instance, the head extended backwards posture, commonly encountered in many routine activities (such as looking upward when negotiating stairs or when changing a light bulb, looking up toward the top shelf of a cabinet or looking and reaching to a cupboard above eye ∗
Corresponding author. Tel.: +33 4 79 75 88 08; fax: +33 4 79 75 81 48. E-mail address: [email protected] (N. Vuillerme).
0966-6362/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.gaitpost.2004.04.007
level), is known to represent a challenging task for the postural system to adapt to (e.g., [2–8]). Clinical reports even indicate that injurious falls are associated with head extension in daily activity [9]. Another example was provided by Endo et al. [10] who concluded that the hyperextended head position during hair shampoo treatment in beauty parlor triggers episodes of vertigo and dizziness and may be a factor for back lifting. In most of the investigations cited above, however, the trajectory of the centre of pressure (CP) represented the studied variable. In fact, this CP assumes two distinct tasks: it counteracts the centre of gravity (CG) in its falling motion and makes it remain in a particular zone within the base of support. In other words, the CP displacements are aimed at facilitating the CG motions to return to a position more compatible with equilibrium, in the first case, and reducing the CG motions as much as possible, in the second case. From this,
N. Vuillerme, P. Rougier / Gait and Posture 21 (2005) 318–325
it seems coherent to dissociate the CP into two elementary components, the centre of gravity vertical projection (CGv ) and the difference between the latter and the former CP − CGv [11]. Biomechanically, the CG motions represent the net performance resulting from CP displacements [12]. This CG, which corresponds to the barycentre of the centres of mass from the different segments, is also thought to be the controlled variable in postural control in humans [13–15]. Complementarily, the difference CP − CGv demonstrates proportionality with the horizontal CG acceleration as long as upright standing can be modelled as an inverted pendulum [16] and constitutes a good expression of the ankle joint stiffness [12,17]. Recently, models such as fractional Brownian motion (fBm) have been applied to gain further insight into the control mechanisms used for maintaining equilibrium. Although this framework has been subject to controversies (e.g., [18,19]), through this model, however, the temporal organisation of various control mechanisms involved in controlling undisturbed upright stance can be revealed in the sense that two distinct control mechanisms, persistent and anti-persistent operate in continual succession. This model also allows the determination of the spatiotemporal characteristics of the switch between these successive mechanisms, or, in other words, from when and to what extent the corrective process begins to operate, on average. Initially used for studying CP trajectories (e.g., [20–24]), this model has recently been extended to the elementary CGv and CP − CGv motions [11] and has been used in the past to specify postural behaviour induced by a prolonged loss of vision [25], by eyelid closure [26], by visual feedback effects [27–28], or by body leaning [29]. Overall, the above experiments evidence that partially deterministic controls successively operate initially on CP − CGv and then on CGv motions through persistent and anti-persistent mechanisms, respectively. In other words, the CP motions have a tendency during the shortest time intervals (t) to drift away from the CG, inferring an increased difference CP − CGv , whereas the CGv during the longest t tends to return to an equilibrium point. The purpose of the present experiment is thus to investigate the effects of head extension on undisturbed upright stance control in humans, by comparing its postural effects to a neutral head reference condition. Considering the dominant role played by vision on postural control (e.g., [30]), the experiment was performed in the absence of visual information in order to better isolate those effects. In addition, the application of the fractional Brownian motion framework to elementary motions, i.e. CP − CGv and CGv , should indicate to which extent the increased sways, generally observed with extended head position in healthy individuals (e.g., [2–7]), is due to (1) a modification of the respective contributions of CP − CGv and/or CGv motions in the global CP trajectories on the one hand and/or (2) the subjects’ ability to detect and control their CGv motions in a less precise manner on the other hand.
319
2. Methods 2.1. Experimental procedure Sixteen young healthy adults (2 women, 14 men; mean age = 24.6 ± 3.5 years; mean body weight = 71.8 ± 10.1 kg; mean height = 179.1 ± 7.1 cm) voluntarily participated in the experiment. They gave their informed consent to the experimental procedure as required by the Helsinki declaration (1964) and the local Ethics Committee. None of the subjects presented any history of motor problem, neurological disease, vestibular impairment or had suffered head trauma or whiplash injury. Eyes closed, the subjects stood barefoot, feet together, their arms hanging loosely by their sides on a triangular force platform (Equi+ model PF01) and were asked to sway as little as possible. The signals issued from the load cells, on which the plate lays, were amplified and converted from analogue to digital form before being recorded on a personal computer. The CP trajectories were then processed, as seen above, in different ways through a specific software program (Equi+−Prog01). In the co-ordinate system used, ML and AP characterise medio-lateral and anteroposterior directions, respectively. Two experimental conditions were presented in a counterbalanced order. In the Neutral head condition, the subjects were asked to keep their gaze in a straight-ahead direction. In the Extended head condition, they were asked to tilt their head backward for at least 50◦ in the sagittal plane. No other movement of the trunk was allowed. The experimenter always stood by the subjects to monitor their posture and their head position throughout the trial. Subjects were asked to adopt the required posture and to stabilise their body sway. Ten seconds later, the sampling was initiated. Each experimental condition included five trials of 64 s with a 64 Hz sample frequency [31]. Rest periods of a similar duration and of about 10 min were allowed between each trial and each condition, respectively, and automatically managed by the recording program. In order to assess possible effects resulting from body inclination [29], the mean positions were also measured along both ML and AP directions.
2.2. CGv and CP − CGv motions’ estimation As stated above, displacements of the CP were split into two elementary components: the vertical projection of the CG (CGv ) and the difference between CP and CGv (CP − CGv ). The determination of the CGv motions was directly derived from the CP trajectories on a frequency basis through an amplitude ratio between CGv and CP [32]. The different steps leading to the CGv and CP − CGv motions estimation have been detailed and illustrated by Fig. 1 in the Rougier and Caron study [11].
320
N. Vuillerme, P. Rougier / Gait and Posture 21 (2005) 318–325
controlled: the closer the scaling regimes are to 0.5, the lesser the control. In addition, depending on whether H is superior or inferior to the 0.5 threshold, persistent (the point is drifting away) or anti-persistent behaviours (the point retraces its steps) can be revealed, respectively. The different steps necessary in this data analysis have been detailed and illustrated by Fig. 2 from the previous reports of Rougier and Farenc [25] and Rougier et al. [29]. It is important to note that such experiments classically reveal that two successive portions are distinguishable in both ML and AP directions for CGv and CP − CGv motions, indicating that a somewhat deterministic (persistent or anti-persistent)
Fig. 1. Mean and standard deviation of the mean velocity (A, C) and the surface covered by the trajectory of CP − CGv and CGv motions (B, D) for the two Neutral and Extended head positions. The two experimental conditions are presented with different symbols: Neutral (white bars) and Extended (black bars) head positions. The significant P-values for comparison between Neutral and Extended head positions also are reported (∗ P < 0.05; ∗∗ P < 0.01; ∗∗∗ P < 0.001).
2.3. Signal processing Three approaches have been adopted to study the CGv and CP − CGv elementary motions: (1) A time-domain analysis first includes the calculation of the surface (mm2 ) covered by the trajectory with a 90% confidence interval [33] and the mean velocity (mm/s) of the CGv and CP − CGv elementary motions. (2) A frequency analysis is based on parameters (root mean square RMS and mean frequency MF) calculated from mean spectral decompositions on specific bandwidths (0–0.5 Hz for CGv and 0–3 Hz for CP − CGv ). (3) A mathematical model termed fractional Brownian motion (fBm), as described initially by Mandelbrot and van Ness [34] whose principle is to enable the assessment of the degree to which a trajectory is controlled. This degree is indeed appreciated through the half-slope of a variogram expressing the mean square displacements x2 as a function of increasing time intervals t. A median value of 0.5 for this half-slope, through which the scaling exponent H is computed, indicates a lack of correlation between past and future increments and suggests a complete lack of control. On the other hand, i.e. if H differs from 0.5, positive (H > 0.5) or negative (H < 0.5) correlations can be inferred, which is indicative of a given part of determinism of the control. Depending on how H is positioned with respect to the median value 0.5, it can be inferred that the trajectory is more or less
Fig. 2. Mean and standard deviation of the rms along the ML (A, E) and AP (B, F) directions and mean frequencies along the ML (C, G) and AP (D, H) directions for CP − CGv and CGv motions and for the two Neutral and Extended head positions. The two experimental conditions are presented with different symbols: Neutral (white bars) and Extended (black bars) head positions. The significant P-values for comparison between Neutral and Extended head positions also are reported (∗ P < 0.05; ∗∗ P < 0.01; ∗∗∗ P < 0.001).
N. Vuillerme, P. Rougier / Gait and Posture 21 (2005) 318–325
scaling regime precedes or succeeds a completely stochastic one. Thus, for each of the two elementary motions and each ML and AP direction, two scaling exponents, indexed as short (Hsl ) and long latencies (Hll ), as well as the co-ordinates of the transition point (t and x2 ) were calculated. 2.4. Statistical analysis Data from both Neutral and Extended head conditions were compared through a non-parametric statistical analysis, a Wilcoxon t-test, which was applied to the parameters derived from the time-domain analysis, the frequency analysis and the fBm modelling. Level of significance was set at 0.05.
321
3.3. Parameters issued from the fBm modelling 3.3.1. Transition point co-ordinates Analysis of the time interval t of the transition point does not show any significant difference, neither for the CP − CGv motions (T = 29, P > 0.05, Fig. 3A and T = 62, P > 0.05, Fig. 3B, for ML and AP directions, respectively), nor for the CGv motions (T = 29, P > 0.05, Fig. 3E and T = 62, P > 0.05, Fig. 3F, for ML and AP directions, respectively). Conversely, significant changes of the spatial characteristic x2 of the transition point for both CP − CGv and CGv motions are observed. On the one hand, analysis of the
3. Results Before presenting the different results, it is important to emphasise that no statistical differences have been found regarding the mean positioning of the CP between the two conditions along both ML and AP directions (T = 40, P > 0.05 and T = 40, P > 0.05, respectively). Thus, data obtained in the Extended head position are not likely to be confounded by a possible effect of leaning postures [29].
3.1. Time-domain parameters As illustrated in Fig. 1, the Extended head condition yields an increased mean velocity of both the CP − CGv and CGv motions (T = 3, P < 0.01, Fig. 1A and T = 15, P < 0.001, Fig. 1C, respectively) and an increased surface covered by the trajectory of these elementary motions (T = 18, P < 0.001, Fig. 1B and T = 2, P < 0.01, Fig. 1D, respectively), relative to the Neutral head condition.
3.2. Frequency parameters Analysis of the CP − CGv motions shows larger RMS in the Extended that Neutral head condition along both ML and AP directions (T = 3, P < 0.01, Fig. 2A and T = 0, P < 0.001, Fig. 2B, respectively), whilst analysis of the MF does not exhibit any significant changes (T = 30.5, P > 0.05, Fig. 2C and T = 61, P > 0.05, Fig. 2D, for ML and AP directions, respectively). Similar results are observed for the CGv motions, the Extended head condition yielding increased RMS relative to the Neutral head condition, along both ML and AP directions (T = 5, P < 0.01, Fig. 2E and T = 21, P < 0.05, Fig. 2F, respectively), and non significant changes for the MF (T = 50, P > 0.05, Fig. 2G and T = 36, P > 0.05, Fig. 2H, for ML and AP directions, respectively).
Fig. 3. Mean and standard deviation of the temporal co-ordinates of the transition point (t) along the ML (A, E) and AP (B, F) directions and the spatial co-ordinates (x2 ) of the transition point along the ML (C, G) and AP (D, H) directions obtained from the fBm modelling for CP − CGv and CGv motions and for the two Neutral and Extended head positions. The two experimental conditions are presented with different symbols: Neutral (white bars) and Extended (black bars) head positions. The significant P-values for comparison between Neutral and Extended head positions also are reported (∗ P< 0.05; ∗∗ P < 0.01; ∗∗∗ P < 0.001).
322
N. Vuillerme, P. Rougier / Gait and Posture 21 (2005) 318–325
CP − CGv motions shows significantly larger x2 in the Extended than Neutral head condition along the AP direction (T = 0, P < 0.001, Fig. 3D), whilst no significant changes are observed along the ML direction (T = 35, P > 0.05, Fig. 3C). Analysis of the CGv motions, on the other hand, shows significantly larger x2 in the Extended than Neutral head condition, along both ML and AP directions (T = 17, P < 0.01, Fig. 3G and T = 20, P < 0.05, Fig. 3H, respectively). 3.3.2. Scaling regimes Analysis of the CP − CGv motions shows smaller short latency scaling regimes Hsl in the Extended than Neutral head condition, along the ML direction (T = 18, P < 0.05, Fig. 4A), whilst no significant changes are observed in the AP direction (T = 55, P > 0.05, Fig. 4B). This suggests that the Extended head position leads the subjects to lessen the control of their CP − CGv motions in the ML direction during the shortest time intervals. Analysis of the CGv motions shows smaller long latency scaling regimes Hll in the Extended than Neutral head condition, along both ML and AP directions (T = 28, P < 0.05, Fig. 4C and T = 27, P < 0.05, Fig. 4D, respectively). This suggests an increased anti-persistent behaviour in the longterm region during the longest time intervals in the Extended relative to the Neutral head condition. In other words, in the Extended head condition, there is an increased probability that CGv movements away from a relative equilibrium point will be offset by corrective adjustments back towards the equilibrium position.
Finally, for both experimental conditions, the results of the long latency scaling exponents Hll for CP − CGv motions and those of shortest latency Hsl for CGv motions are close to 0.5, hence indicating a behaviour solely stochastic in nature. 4. Discussion The purpose of the present experiment was to investigate the effects of head extension on undisturbed upright stance control in humans. To this aim, 16 young healthy adults were asked to stand as immobile as possible in two Neutral and Extended head conditions, with their eyes closed. Overall, the results confirm previous investigations reporting a deterioration of postural control with extended head position in young healthy adults (e.g., [2–7]). However, further information is given with regard to the variable affected by the extended head position. It is indeed worth re-emphasising that the CP is a complex signal, hence justifying the decomposition into two elementary CP − CGv and CGv motions. 4.1. Head extension effects on CP − CGv motions The Extended head position yields larger CP − CGv amplitudes, as indicated by the mean velocity (Fig. 1A), surface (Fig. 1B) and RMS along both ML and AP direction (Fig. 2A and B), whilst the frequency distributions remain unchanged (Fig. 2C and D). From a biomechanical point of view, increasing the amplitudes of CP − CGv motions in the Extended head condition, seen as an expression of
Fig. 4. Mean and standard deviation of the short latency scaling exponents (Hsl ) for the CP − CGv motions and the long latency scaling exponents (Hll ) along the ML (A and C) and AP (B and D) directions for the CGv motions obtained from the fBm modelling and for the two Neutral and Extended head positions. The two experimental conditions are presented with different symbols: Neutral (white bars) and Extended (black bars) head positions. The significant P-values for comparison between Neutral and Extended head positions also are reported (∗ P < 0.05).
N. Vuillerme, P. Rougier / Gait and Posture 21 (2005) 318–325
the initial horizontal acceleration communicated to the CGv [12,16], negatively affects the relative facility for the subjects to handle CGv motions in this condition due to the higher forces they would have to counteract. With regard to the hypothesis of an impairment of cervical proprioceptive information caused by the Extended head position, these results are in line with those of Ivanenko et al. [35]. Indeed, increases in the ankle torque and in the activity of the soleus muscle is observed when the reliability of the cervical proprioceptive information is altered by vibration of neck muscles [35]. In this case, neck muscle vibration was specifically used to activate the muscle spindle primary endings (e.g., [36]) and to modify the perception of body orientation (e.g., [37,38]). It also is possible that the observed changes CP − CGv motions to be induced by the tonic vestibular and neck reflexes (e.g., [39]), the Extended head position increasing stiffness of the lower limbs by enhancing the level of muscular activity across the knee and ankle joints. However, it is also worth noting that these observed postural effects are not limited to a sensory perturbation of cervical origin. Similar increased level of muscular activity at the ankle level, expressed through the amplitudes of the CP − CGv motions, are rather observed when postural control is challenged by the suppression of vision [26] or the alteration or loss of somatosensory information from the lower limbs resulting from disease (e.g., diabetic peripheral neuropathy [40]). More largely, with respect to the postural instability induced by the Extended head posture, the increased CP − CGv motions observed in the present experiment are consistent with the results of a recent study investigating the mechanisms contributing to age-related increases in postural sway and falls in the elderly [41]. These authors suggest that high levels of muscle activity across the lower limb joints, expressed through electromyographic data, may be characteristic of age-related declines in postural control and that such exaggerated activity may be responsible for the enlarged short-term postural sway and hence may compromise the older adults’ ability to control undisturbed upright stance. Complementarily to the time-domain and frequency analyses, the fBm modelling provides additional insight into the nature and the temporal organisation of the control mechanisms called into play in the Extended head condition. Indeed, the co-ordinates of the transition points present some statistically significant changes for the two experimental conditions (Fig. 3). In fact, the observed differences occur for the spatial co-ordinates along the AP direction, with larger values in the Extended than Neutral head condition (Fig. 3D). A similar feature is also perceptible along the ML direction, although without significant effect (Fig. 3C). Thus, it follows from these results that the increased stiffness observed in the Extended head condition stems from an enhanced transition point mean square distances characterising CP − CGv motions. It is important to recall that the augmented spatial threshold x2 of the transition point for the CP − CGv motions observed in the Extended head
323
condition, determining larger initial horizontal accelerations communicated to the CG [12,16], worsen the control of the CG. 4.2. Head extension effects on CGv motions As for the CP − CGv motions, the Extended head condition yields larger displacements of the CGv , as indicated by the mean velocity (Fig. 1C), surface (Fig. 1D) and RMS along both ML and AP direction (Fig. 2E and F), whilst the frequency distributions remain unchanged (Fig. 2G and H). Modelling CGv motions as a fractional Brownian motion process provides further insight. On the one hand, the observed increased rms in the Extended head condition is likely to be related to spatial parameters, since augmented spatial co-ordinates of the transition points were observed in the Extended head condition, along both ML and AP directions (Fig. 3G and H, respectively). These larger distances covered by the CG before a corrective mechanism begins to operate could also be explained by the impairment of sensory perception and integration caused by the extended head position. Indeed, tipping the head backwards, modifies de facto the head position with respect to the body and to the space, this information being derived from the sensory inputs of the neck proprioceptors and the otoliths, respectively. Whether the increased CGv motions observed in the Extended head condition are due to abnormal sensory inputs arising from neck proprioceptors (e.g., [4]), change in the orientation of the vestibular organs that may place the utricular otoliths well beyond their working range (e.g., [3,4]) and/or central integrative functions is rather difficult to answer [6] and is not within the scope of our study. However, with regard to the hypothesis of a cervical proprioceptive perturbation induced by the Extended head posture, our results confirm the important role of proprioceptive mechanisms of the neck region in postural control. Indeed, postural control decreases when the reliance of cervical proprioceptive information is impaired by pathology, trauma or injury (e.g., [5,42,43]) or experimental manipulations in normal subjects, such as neck muscles vibration (e.g., [36,44,45]). On the other hand, the lack of modification of the MF means, by definition, that no change of the period needed for the CGv to return to a similar position, occurs between the two experimental conditions. Considering (1) the increased distance covered by the CG before a corrective mechanism intervenes observed in the Extended head condition (Fig. 3G and H) and (2) the diminished long latency scaling exponent Hll associated with (Fig. 4C and D), this result would mainly stem from a larger contribution of deterministic processes in these corrective mechanisms. These more controlled CGv motions, especially those aimed at making the CG return more directly to its initial position, may then be considered as an adaptive process to the larger amount of sway and the increased spatial co-ordinates of the transition points. This hypothesis is in accordance with previous results. Indeed, an
324
N. Vuillerme, P. Rougier / Gait and Posture 21 (2005) 318–325
enhanced control in the correction of the CGv motions associated with an increased spatial threshold were observed when manipulating the characteristics of the subjects (e.g., [22]), the visual environment [26], the availability of visual feedback [27,28] or the standing postures [29]. Such a process, however, appears ineffective at limiting body sway in the Extended head position. In conclusion, in light of the CP displacements dissociation into two elementary CGv and CP − CGv motions and the recourse to parameters issued from frequency analysis and fBm modelling, the present findings evidence that the enlarged territories covered by the CP trajectories, usually observed with extended head posture, are the result of an exaggerated muscular activity and lessened capacities to detect and control the CG motions. It is concluded that the head extended posture commonly encountered in many routine activities, may represent a challenge for the postural control system, even for young healthy adults. Whether the present effects can again be observed in people showing less accurate postural capacities (e.g., elderly persons) for whom the consequences of an impaired postural control could be more dramatic remains to be investigated and is included in our immediate plans. Finally, although the extended head posture probably impacts on various postural tasks, the nature of these impacts is yet to be established. Whether and how such an extended head posture may modulate other postural control mechanisms, such as anticipatory adjustments for voluntary movements or automatic postural reactions to external perturbations, is currently being investigated.
Acknowledgements The authors would like to thank the anonymous reviewers for helpful comments and suggestions. This paper was written while the first author was ATER at Universit´e de Savoie, France.
References [1] Newell KM. Constraints on the development of coordination. In: Wade MG, Whiting HTA, editors. Motor development in children: aspects of coordination and control. Boston: Martinus Nijhoff; 1986. p. 341– 60. [2] Barin K, Seitz CM, Welling DB. Effect of head orientation on the diagnostic sensitivity of posturography in patients with compensated unilateral lesions. Otolaryngol Head Neck Surg 1992;106:355–62. [3] Brandt T, Krafczyk S, Malsbenden I. Postural imbalance with head extension: improvement by training as a model for ataxia therapy. Ann N Y Acad Sci 1981;374:636–49. [4] Jackson RT, Epstein CM. Effect of head extension on equilibrium in normal subjects. Ann Otol Rhinol Laryngol 1991;100:63–7. [5] Kogler A, Lindfors J, Odkvist LM, Ledin T. Postural stability using different neck positions in normal subjects and patients with neck trauma. Acta Otolaryngol 2000;120:151–5. [6] Norr´e ME. Head extension effect in static posturography. Ann Otol Rhinol Laryngol 1995;104:570–3.
[7] Norr´e ME, Forrez G. Posture testing (posturography) in the diagnosis of peripheral vestibular pathology. Arch Otorhinolaryngol 1986;243:186–9. [8] Simoneau GG, Leibowitz HW, Ulbrecht JS, Tyrrell RA, Cavanagh PR. The effects of visual factors and head orientation on postural steadiness in women 55 to 70 years of age. J Gerontol 1992;47:M151–8. [9] Sheldon JH. On the natural history of falls in old age. Br Med J 1960;2:1685–90. [10] Endo K, Ichimaru K, Shimura H, Imakiire A. Cervical vertigo after hair shampoo treatment at a hairdressing salon: a case report. Spine 2000;25:632–4. [11] Rougier P, Caron O. Centre of gravity motions and ankle joint stiffness control in upright undisturbed stance modelled through fractional Brownian motion framework. J Mot Behav 2000;32:405–13. [12] Winter DA, Patla AE, Prince F, Ishac M, Gielo-Perczak K. Stiffness control of balance in quiet standing. J Neurophysiol 1998;80:1211–21. [13] Cl´ement G, Gurfinkel VS, Lestienne F, Lipshits MI, Popov KE. Adaptation of postural control to weightlessness. Exp Brain Res 1984;57:61–72. [14] Horstmann A, Dietz V. A basic posture control mechanism: the stabilization of the centre of gravity. Electroenceph Clin Neurophysiol 1990;76:165–76. [15] Winter DA, Prince F, Franck JS, Powel C, Zabjek KF. Unified theory regarding A/P and M/L balance in quiet standing. J Neurophysiol 1996;75:2334–43. [16] Breni`ere Y, Do MC, Bouisset S. Are dynamic phenomena prior to stepping essential to walking? J Mot Behav 1987;19:62–76. [17] Caron O, G´elat T, Rougier P, Blanchi JP. A comparative analysis of the centre of gravity and centre of pressure sway paths in standing posture: an estimation of active stiffness. J Appl Biomech 2000;16:234–47. [18] Deligni`eres D, Deschamps T, Legros A, Caillou N. A methodological note on nonlinear time series analysis: is the open- and closedloop model of Collins and De Luca (1993) a statistical artifact? J Mot Behav 2003;35:86–97. [19] Peterka RJ. Postural control model interpretation of stabilogram diffusion analysis. Biol Cybern 2000;82:335–43. [20] Chiari L, Cappello A, Lenzi D, Della Croce U. An improved technique for the extraction of stochastic parameters from stabilograms. Gait Posture 2000;12:225–34. [21] Collins JJ, De Luca CJ. Open-loop and closed-loop control of posture: a random-walk analysis of centre-of-pressure trajectories. Exp Brain Res 1993;95:308–18. [22] Collins JJ, De Luca CJ. The effects of visual input on openloop and closed-loop postural control mechanisms. Exp Brain Res 1995;103:151–63. [23] Riley MA, Wong S, Mitra S, Turvey MT. Common effects of touch and vision on postural parameters. Exp Brain Res 1997;117:165–70. [24] Rougier P. Influence of visual feedback on successive control mechanisms in upright stance in humans assessed by fractional Brownian motion modelling. Neurosci Lett 1999;266:157–60. [25] Rougier P, Farenc I. Adaptative effects of loss of vision on upright undisturbed stance. Brain Res 2000;871:165–74. [26] Rougier P. The influence of having the eyelids open or closed on undisturbed postural control. Neurosci Res 2003;47:73–83. [27] Rougier P. Optimising the visual feedback technique for improving upright stance maintenance by delaying its display: behavioural effects on healthy adults. Gait Posture 2004;19:154–63. [28] Rougier P. Visual feedback induces opposite effects on elementary centre of gravity and centre of pressure minus centre of gravity motions in undisturbed upright stance. Clin Biomech 2003;18:341– 9. [29] Rougier P, Burdet C, Farenc I, Berger L. Backward and forward leaning postures modelled by a fBm framework. Neurosci Res 2001;41:41–50.
N. Vuillerme, P. Rougier / Gait and Posture 21 (2005) 318–325 [30] Paulus WM, Straube A, Brandt T. Visual stabilization of posture: physiological stimulus characteristics and clinical aspects. Brain 1984;107:1143–63. [31] Carpenter MG, Frank JS, Winter DA, Peysar GW. Sampling duration effects on centre of pressure summary measures. Gait Posture 2001;13:35–40. [32] Caron O, Faure B, Breni`ere Y. Estimating the center of gravity of the body on the basis of the center of pressure in standing posture. J Biomech 1997;30:1169–71. [33] Tagaki A, Fujimura E, Suehiro S. A new method of statokinesigram area measurement. Application of a statistically calculated ellipse. In: Igarashi M, Black O, editors. Vestibular and visual control on posture and locomotor equilibrium. Bˆale: Karger; 1985. p. 74–9. [34] Mandelbrot BB, van Ness JW. Fractional Brownian motions, fractional noises and applications. SIAM Rev 1968;10:422–37. [35] Ivanenko YP, Talis VL, Kazennikov OV. Support stability influences postural responses vibrations in humans. Eur J Neurosci 1999;11:647–54. [36] Roll JP, Vedel JP, Ribot E. Alteration of proprioceptive messages induced by tendon vibration in man: a microneurographic study. Environ Brain Res 1989;76:213–22. [37] Karnath HO, Sievering D, Fetter M. The interactive contribution of neck muscle proprioception and vestibular stimulation to subjective “straight ahead” orientation in man. Exp Brain Res 1994;101:140– 6.
325
[38] Strupp M, Arbusow V, Borges Pereira C, Dieterich M, Brandt T. Subjective straight-ahead during neck muscle vibration: effects of ageing. NeuroReport 1999;10:3191–4. [39] Tokizane T, Murao M, Ogata T, Kondo T. Electromyographic studies on tonic neck, lumbar, and labyrinthine reflexes in normal persons. Jpn J Physiol 1951;2:30–146. [40] Corriveau H, Prince F, Hebert R, Raiche M, Tessier D, Maheux P, et al. Evaluation of postural stability in elderly with diabetic neuropathy. Diabetes Care 2000;23:1187–91. [41] Laughton CA, Slavin M, Katdare K, Nolan L, Bean JF, Kerrigan DC, et al. Aging, muscle activity, and balance control: physiologic changes associated with balance impairment. Gait Posture 2003;180:101–8. [42] Karlberg M, Persson L, Magnusson M. Reduced postural control in patients with chronic cervicobrachial pain syndrome. Gait Posture 1995;3:241–9. [43] Michaelson P, Michaelson M, Jaric S, Latash ML, Sj¨olander P, Djupsj¨obacka M. Vertical posture and head stability in patients with chronic neck pain. J Rehabil Med 2003;35:229–35. [44] Gregoric M, Tayeka T, Baron JB, Bessineton JC. Influence of vibration of neck muscles on balance control in man. Agressologie 1978;19:37–8. [45] Koskimies K, Sutinen P, Aalto H, Starck J, Toppila E, Hirvonen T, et al. Postural stability, neck proprioception and tension neck. Acta Otolaryngol Suppl 1997;529:95–7.
Effects of head extension on undisturbed upright stance control in humans Nicolas Vuillerme∗ , Patrice Rougier Laboratoire de Mod´elisation des Activit´es Sportives, Universit´e de Savoie, Domaine Universitaire, 73376 Le Bourget du Lac cedex, France Received 17 November 2003; accepted 1 April 2004
Abstract The purpose of the present experiment was to investigate whether and how the head extended posture, commonly encountered in many routine activities, affects undisturbed upright stance control mechanisms in humans. Sixteen young healthy adults stood feet together, with their eyes closed and were asked to sway as little as possible in two Neutral and Extended head conditions. Centre of pressure (CP) displacements, recorded using a force platform, were used to compute the motions of the vertical projection of the centre of gravity (CGv ) and those of the difference CP − CGv . A time-domain analysis shows increased mean velocity and surface covered by the trajectory of both elementary motions in the Extended head condition. A frequency analysis also reveals increased root mean squares on the CP − CGv motions, suggesting increased muscular activity in the Extended head condition. Furthermore, similar changes occur on CGv motions. Finally, modelling these trajectories as a fractional Brownian motion process demonstrates increased spatial transition point co-ordinates at which the corrective process is initiated and a more deterministic control mechanism in this corrective process involving CGv motions in the Extended head position. Together, the present findings suggest that head extension position yields a reorganisation of the control mechanisms for maintaining undisturbed upright stance. © 2004 Elsevier B.V. All rights reserved. Keywords: Undisturbed upright stance control; Head position; Centre of pressure; Centre of gravity; Frequency analysis; Fractional Brownian motion modelling
1. Introduction The ability to maintain an upright stance is essential in gait and in the initiation of voluntary movements that are vital not only for professional, sportive but also for daily living activities. As we move in a wide range of situations, under changing tasks and environmental demands, postural control entails adapting to various task parameters (e.g., [1]). However, imbalance might occur under specific situations, conditions and/or postures. For instance, the head extended backwards posture, commonly encountered in many routine activities (such as looking upward when negotiating stairs or when changing a light bulb, looking up toward the top shelf of a cabinet or looking and reaching to a cupboard above eye ∗
Corresponding author. Tel.: +33 4 79 75 88 08; fax: +33 4 79 75 81 48. E-mail address: [email protected] (N. Vuillerme).
0966-6362/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.gaitpost.2004.04.007
level), is known to represent a challenging task for the postural system to adapt to (e.g., [2–8]). Clinical reports even indicate that injurious falls are associated with head extension in daily activity [9]. Another example was provided by Endo et al. [10] who concluded that the hyperextended head position during hair shampoo treatment in beauty parlor triggers episodes of vertigo and dizziness and may be a factor for back lifting. In most of the investigations cited above, however, the trajectory of the centre of pressure (CP) represented the studied variable. In fact, this CP assumes two distinct tasks: it counteracts the centre of gravity (CG) in its falling motion and makes it remain in a particular zone within the base of support. In other words, the CP displacements are aimed at facilitating the CG motions to return to a position more compatible with equilibrium, in the first case, and reducing the CG motions as much as possible, in the second case. From this,
N. Vuillerme, P. Rougier / Gait and Posture 21 (2005) 318–325
it seems coherent to dissociate the CP into two elementary components, the centre of gravity vertical projection (CGv ) and the difference between the latter and the former CP − CGv [11]. Biomechanically, the CG motions represent the net performance resulting from CP displacements [12]. This CG, which corresponds to the barycentre of the centres of mass from the different segments, is also thought to be the controlled variable in postural control in humans [13–15]. Complementarily, the difference CP − CGv demonstrates proportionality with the horizontal CG acceleration as long as upright standing can be modelled as an inverted pendulum [16] and constitutes a good expression of the ankle joint stiffness [12,17]. Recently, models such as fractional Brownian motion (fBm) have been applied to gain further insight into the control mechanisms used for maintaining equilibrium. Although this framework has been subject to controversies (e.g., [18,19]), through this model, however, the temporal organisation of various control mechanisms involved in controlling undisturbed upright stance can be revealed in the sense that two distinct control mechanisms, persistent and anti-persistent operate in continual succession. This model also allows the determination of the spatiotemporal characteristics of the switch between these successive mechanisms, or, in other words, from when and to what extent the corrective process begins to operate, on average. Initially used for studying CP trajectories (e.g., [20–24]), this model has recently been extended to the elementary CGv and CP − CGv motions [11] and has been used in the past to specify postural behaviour induced by a prolonged loss of vision [25], by eyelid closure [26], by visual feedback effects [27–28], or by body leaning [29]. Overall, the above experiments evidence that partially deterministic controls successively operate initially on CP − CGv and then on CGv motions through persistent and anti-persistent mechanisms, respectively. In other words, the CP motions have a tendency during the shortest time intervals (t) to drift away from the CG, inferring an increased difference CP − CGv , whereas the CGv during the longest t tends to return to an equilibrium point. The purpose of the present experiment is thus to investigate the effects of head extension on undisturbed upright stance control in humans, by comparing its postural effects to a neutral head reference condition. Considering the dominant role played by vision on postural control (e.g., [30]), the experiment was performed in the absence of visual information in order to better isolate those effects. In addition, the application of the fractional Brownian motion framework to elementary motions, i.e. CP − CGv and CGv , should indicate to which extent the increased sways, generally observed with extended head position in healthy individuals (e.g., [2–7]), is due to (1) a modification of the respective contributions of CP − CGv and/or CGv motions in the global CP trajectories on the one hand and/or (2) the subjects’ ability to detect and control their CGv motions in a less precise manner on the other hand.
319
2. Methods 2.1. Experimental procedure Sixteen young healthy adults (2 women, 14 men; mean age = 24.6 ± 3.5 years; mean body weight = 71.8 ± 10.1 kg; mean height = 179.1 ± 7.1 cm) voluntarily participated in the experiment. They gave their informed consent to the experimental procedure as required by the Helsinki declaration (1964) and the local Ethics Committee. None of the subjects presented any history of motor problem, neurological disease, vestibular impairment or had suffered head trauma or whiplash injury. Eyes closed, the subjects stood barefoot, feet together, their arms hanging loosely by their sides on a triangular force platform (Equi+ model PF01) and were asked to sway as little as possible. The signals issued from the load cells, on which the plate lays, were amplified and converted from analogue to digital form before being recorded on a personal computer. The CP trajectories were then processed, as seen above, in different ways through a specific software program (Equi+−Prog01). In the co-ordinate system used, ML and AP characterise medio-lateral and anteroposterior directions, respectively. Two experimental conditions were presented in a counterbalanced order. In the Neutral head condition, the subjects were asked to keep their gaze in a straight-ahead direction. In the Extended head condition, they were asked to tilt their head backward for at least 50◦ in the sagittal plane. No other movement of the trunk was allowed. The experimenter always stood by the subjects to monitor their posture and their head position throughout the trial. Subjects were asked to adopt the required posture and to stabilise their body sway. Ten seconds later, the sampling was initiated. Each experimental condition included five trials of 64 s with a 64 Hz sample frequency [31]. Rest periods of a similar duration and of about 10 min were allowed between each trial and each condition, respectively, and automatically managed by the recording program. In order to assess possible effects resulting from body inclination [29], the mean positions were also measured along both ML and AP directions.
2.2. CGv and CP − CGv motions’ estimation As stated above, displacements of the CP were split into two elementary components: the vertical projection of the CG (CGv ) and the difference between CP and CGv (CP − CGv ). The determination of the CGv motions was directly derived from the CP trajectories on a frequency basis through an amplitude ratio between CGv and CP [32]. The different steps leading to the CGv and CP − CGv motions estimation have been detailed and illustrated by Fig. 1 in the Rougier and Caron study [11].
320
N. Vuillerme, P. Rougier / Gait and Posture 21 (2005) 318–325
controlled: the closer the scaling regimes are to 0.5, the lesser the control. In addition, depending on whether H is superior or inferior to the 0.5 threshold, persistent (the point is drifting away) or anti-persistent behaviours (the point retraces its steps) can be revealed, respectively. The different steps necessary in this data analysis have been detailed and illustrated by Fig. 2 from the previous reports of Rougier and Farenc [25] and Rougier et al. [29]. It is important to note that such experiments classically reveal that two successive portions are distinguishable in both ML and AP directions for CGv and CP − CGv motions, indicating that a somewhat deterministic (persistent or anti-persistent)
Fig. 1. Mean and standard deviation of the mean velocity (A, C) and the surface covered by the trajectory of CP − CGv and CGv motions (B, D) for the two Neutral and Extended head positions. The two experimental conditions are presented with different symbols: Neutral (white bars) and Extended (black bars) head positions. The significant P-values for comparison between Neutral and Extended head positions also are reported (∗ P < 0.05; ∗∗ P < 0.01; ∗∗∗ P < 0.001).
2.3. Signal processing Three approaches have been adopted to study the CGv and CP − CGv elementary motions: (1) A time-domain analysis first includes the calculation of the surface (mm2 ) covered by the trajectory with a 90% confidence interval [33] and the mean velocity (mm/s) of the CGv and CP − CGv elementary motions. (2) A frequency analysis is based on parameters (root mean square RMS and mean frequency MF) calculated from mean spectral decompositions on specific bandwidths (0–0.5 Hz for CGv and 0–3 Hz for CP − CGv ). (3) A mathematical model termed fractional Brownian motion (fBm), as described initially by Mandelbrot and van Ness [34] whose principle is to enable the assessment of the degree to which a trajectory is controlled. This degree is indeed appreciated through the half-slope of a variogram expressing the mean square displacements x2 as a function of increasing time intervals t. A median value of 0.5 for this half-slope, through which the scaling exponent H is computed, indicates a lack of correlation between past and future increments and suggests a complete lack of control. On the other hand, i.e. if H differs from 0.5, positive (H > 0.5) or negative (H < 0.5) correlations can be inferred, which is indicative of a given part of determinism of the control. Depending on how H is positioned with respect to the median value 0.5, it can be inferred that the trajectory is more or less
Fig. 2. Mean and standard deviation of the rms along the ML (A, E) and AP (B, F) directions and mean frequencies along the ML (C, G) and AP (D, H) directions for CP − CGv and CGv motions and for the two Neutral and Extended head positions. The two experimental conditions are presented with different symbols: Neutral (white bars) and Extended (black bars) head positions. The significant P-values for comparison between Neutral and Extended head positions also are reported (∗ P < 0.05; ∗∗ P < 0.01; ∗∗∗ P < 0.001).
N. Vuillerme, P. Rougier / Gait and Posture 21 (2005) 318–325
scaling regime precedes or succeeds a completely stochastic one. Thus, for each of the two elementary motions and each ML and AP direction, two scaling exponents, indexed as short (Hsl ) and long latencies (Hll ), as well as the co-ordinates of the transition point (t and x2 ) were calculated. 2.4. Statistical analysis Data from both Neutral and Extended head conditions were compared through a non-parametric statistical analysis, a Wilcoxon t-test, which was applied to the parameters derived from the time-domain analysis, the frequency analysis and the fBm modelling. Level of significance was set at 0.05.
321
3.3. Parameters issued from the fBm modelling 3.3.1. Transition point co-ordinates Analysis of the time interval t of the transition point does not show any significant difference, neither for the CP − CGv motions (T = 29, P > 0.05, Fig. 3A and T = 62, P > 0.05, Fig. 3B, for ML and AP directions, respectively), nor for the CGv motions (T = 29, P > 0.05, Fig. 3E and T = 62, P > 0.05, Fig. 3F, for ML and AP directions, respectively). Conversely, significant changes of the spatial characteristic x2 of the transition point for both CP − CGv and CGv motions are observed. On the one hand, analysis of the
3. Results Before presenting the different results, it is important to emphasise that no statistical differences have been found regarding the mean positioning of the CP between the two conditions along both ML and AP directions (T = 40, P > 0.05 and T = 40, P > 0.05, respectively). Thus, data obtained in the Extended head position are not likely to be confounded by a possible effect of leaning postures [29].
3.1. Time-domain parameters As illustrated in Fig. 1, the Extended head condition yields an increased mean velocity of both the CP − CGv and CGv motions (T = 3, P < 0.01, Fig. 1A and T = 15, P < 0.001, Fig. 1C, respectively) and an increased surface covered by the trajectory of these elementary motions (T = 18, P < 0.001, Fig. 1B and T = 2, P < 0.01, Fig. 1D, respectively), relative to the Neutral head condition.
3.2. Frequency parameters Analysis of the CP − CGv motions shows larger RMS in the Extended that Neutral head condition along both ML and AP directions (T = 3, P < 0.01, Fig. 2A and T = 0, P < 0.001, Fig. 2B, respectively), whilst analysis of the MF does not exhibit any significant changes (T = 30.5, P > 0.05, Fig. 2C and T = 61, P > 0.05, Fig. 2D, for ML and AP directions, respectively). Similar results are observed for the CGv motions, the Extended head condition yielding increased RMS relative to the Neutral head condition, along both ML and AP directions (T = 5, P < 0.01, Fig. 2E and T = 21, P < 0.05, Fig. 2F, respectively), and non significant changes for the MF (T = 50, P > 0.05, Fig. 2G and T = 36, P > 0.05, Fig. 2H, for ML and AP directions, respectively).
Fig. 3. Mean and standard deviation of the temporal co-ordinates of the transition point (t) along the ML (A, E) and AP (B, F) directions and the spatial co-ordinates (x2 ) of the transition point along the ML (C, G) and AP (D, H) directions obtained from the fBm modelling for CP − CGv and CGv motions and for the two Neutral and Extended head positions. The two experimental conditions are presented with different symbols: Neutral (white bars) and Extended (black bars) head positions. The significant P-values for comparison between Neutral and Extended head positions also are reported (∗ P< 0.05; ∗∗ P < 0.01; ∗∗∗ P < 0.001).
322
N. Vuillerme, P. Rougier / Gait and Posture 21 (2005) 318–325
CP − CGv motions shows significantly larger x2 in the Extended than Neutral head condition along the AP direction (T = 0, P < 0.001, Fig. 3D), whilst no significant changes are observed along the ML direction (T = 35, P > 0.05, Fig. 3C). Analysis of the CGv motions, on the other hand, shows significantly larger x2 in the Extended than Neutral head condition, along both ML and AP directions (T = 17, P < 0.01, Fig. 3G and T = 20, P < 0.05, Fig. 3H, respectively). 3.3.2. Scaling regimes Analysis of the CP − CGv motions shows smaller short latency scaling regimes Hsl in the Extended than Neutral head condition, along the ML direction (T = 18, P < 0.05, Fig. 4A), whilst no significant changes are observed in the AP direction (T = 55, P > 0.05, Fig. 4B). This suggests that the Extended head position leads the subjects to lessen the control of their CP − CGv motions in the ML direction during the shortest time intervals. Analysis of the CGv motions shows smaller long latency scaling regimes Hll in the Extended than Neutral head condition, along both ML and AP directions (T = 28, P < 0.05, Fig. 4C and T = 27, P < 0.05, Fig. 4D, respectively). This suggests an increased anti-persistent behaviour in the longterm region during the longest time intervals in the Extended relative to the Neutral head condition. In other words, in the Extended head condition, there is an increased probability that CGv movements away from a relative equilibrium point will be offset by corrective adjustments back towards the equilibrium position.
Finally, for both experimental conditions, the results of the long latency scaling exponents Hll for CP − CGv motions and those of shortest latency Hsl for CGv motions are close to 0.5, hence indicating a behaviour solely stochastic in nature. 4. Discussion The purpose of the present experiment was to investigate the effects of head extension on undisturbed upright stance control in humans. To this aim, 16 young healthy adults were asked to stand as immobile as possible in two Neutral and Extended head conditions, with their eyes closed. Overall, the results confirm previous investigations reporting a deterioration of postural control with extended head position in young healthy adults (e.g., [2–7]). However, further information is given with regard to the variable affected by the extended head position. It is indeed worth re-emphasising that the CP is a complex signal, hence justifying the decomposition into two elementary CP − CGv and CGv motions. 4.1. Head extension effects on CP − CGv motions The Extended head position yields larger CP − CGv amplitudes, as indicated by the mean velocity (Fig. 1A), surface (Fig. 1B) and RMS along both ML and AP direction (Fig. 2A and B), whilst the frequency distributions remain unchanged (Fig. 2C and D). From a biomechanical point of view, increasing the amplitudes of CP − CGv motions in the Extended head condition, seen as an expression of
Fig. 4. Mean and standard deviation of the short latency scaling exponents (Hsl ) for the CP − CGv motions and the long latency scaling exponents (Hll ) along the ML (A and C) and AP (B and D) directions for the CGv motions obtained from the fBm modelling and for the two Neutral and Extended head positions. The two experimental conditions are presented with different symbols: Neutral (white bars) and Extended (black bars) head positions. The significant P-values for comparison between Neutral and Extended head positions also are reported (∗ P < 0.05).
N. Vuillerme, P. Rougier / Gait and Posture 21 (2005) 318–325
the initial horizontal acceleration communicated to the CGv [12,16], negatively affects the relative facility for the subjects to handle CGv motions in this condition due to the higher forces they would have to counteract. With regard to the hypothesis of an impairment of cervical proprioceptive information caused by the Extended head position, these results are in line with those of Ivanenko et al. [35]. Indeed, increases in the ankle torque and in the activity of the soleus muscle is observed when the reliability of the cervical proprioceptive information is altered by vibration of neck muscles [35]. In this case, neck muscle vibration was specifically used to activate the muscle spindle primary endings (e.g., [36]) and to modify the perception of body orientation (e.g., [37,38]). It also is possible that the observed changes CP − CGv motions to be induced by the tonic vestibular and neck reflexes (e.g., [39]), the Extended head position increasing stiffness of the lower limbs by enhancing the level of muscular activity across the knee and ankle joints. However, it is also worth noting that these observed postural effects are not limited to a sensory perturbation of cervical origin. Similar increased level of muscular activity at the ankle level, expressed through the amplitudes of the CP − CGv motions, are rather observed when postural control is challenged by the suppression of vision [26] or the alteration or loss of somatosensory information from the lower limbs resulting from disease (e.g., diabetic peripheral neuropathy [40]). More largely, with respect to the postural instability induced by the Extended head posture, the increased CP − CGv motions observed in the present experiment are consistent with the results of a recent study investigating the mechanisms contributing to age-related increases in postural sway and falls in the elderly [41]. These authors suggest that high levels of muscle activity across the lower limb joints, expressed through electromyographic data, may be characteristic of age-related declines in postural control and that such exaggerated activity may be responsible for the enlarged short-term postural sway and hence may compromise the older adults’ ability to control undisturbed upright stance. Complementarily to the time-domain and frequency analyses, the fBm modelling provides additional insight into the nature and the temporal organisation of the control mechanisms called into play in the Extended head condition. Indeed, the co-ordinates of the transition points present some statistically significant changes for the two experimental conditions (Fig. 3). In fact, the observed differences occur for the spatial co-ordinates along the AP direction, with larger values in the Extended than Neutral head condition (Fig. 3D). A similar feature is also perceptible along the ML direction, although without significant effect (Fig. 3C). Thus, it follows from these results that the increased stiffness observed in the Extended head condition stems from an enhanced transition point mean square distances characterising CP − CGv motions. It is important to recall that the augmented spatial threshold x2 of the transition point for the CP − CGv motions observed in the Extended head
323
condition, determining larger initial horizontal accelerations communicated to the CG [12,16], worsen the control of the CG. 4.2. Head extension effects on CGv motions As for the CP − CGv motions, the Extended head condition yields larger displacements of the CGv , as indicated by the mean velocity (Fig. 1C), surface (Fig. 1D) and RMS along both ML and AP direction (Fig. 2E and F), whilst the frequency distributions remain unchanged (Fig. 2G and H). Modelling CGv motions as a fractional Brownian motion process provides further insight. On the one hand, the observed increased rms in the Extended head condition is likely to be related to spatial parameters, since augmented spatial co-ordinates of the transition points were observed in the Extended head condition, along both ML and AP directions (Fig. 3G and H, respectively). These larger distances covered by the CG before a corrective mechanism begins to operate could also be explained by the impairment of sensory perception and integration caused by the extended head position. Indeed, tipping the head backwards, modifies de facto the head position with respect to the body and to the space, this information being derived from the sensory inputs of the neck proprioceptors and the otoliths, respectively. Whether the increased CGv motions observed in the Extended head condition are due to abnormal sensory inputs arising from neck proprioceptors (e.g., [4]), change in the orientation of the vestibular organs that may place the utricular otoliths well beyond their working range (e.g., [3,4]) and/or central integrative functions is rather difficult to answer [6] and is not within the scope of our study. However, with regard to the hypothesis of a cervical proprioceptive perturbation induced by the Extended head posture, our results confirm the important role of proprioceptive mechanisms of the neck region in postural control. Indeed, postural control decreases when the reliance of cervical proprioceptive information is impaired by pathology, trauma or injury (e.g., [5,42,43]) or experimental manipulations in normal subjects, such as neck muscles vibration (e.g., [36,44,45]). On the other hand, the lack of modification of the MF means, by definition, that no change of the period needed for the CGv to return to a similar position, occurs between the two experimental conditions. Considering (1) the increased distance covered by the CG before a corrective mechanism intervenes observed in the Extended head condition (Fig. 3G and H) and (2) the diminished long latency scaling exponent Hll associated with (Fig. 4C and D), this result would mainly stem from a larger contribution of deterministic processes in these corrective mechanisms. These more controlled CGv motions, especially those aimed at making the CG return more directly to its initial position, may then be considered as an adaptive process to the larger amount of sway and the increased spatial co-ordinates of the transition points. This hypothesis is in accordance with previous results. Indeed, an
324
N. Vuillerme, P. Rougier / Gait and Posture 21 (2005) 318–325
enhanced control in the correction of the CGv motions associated with an increased spatial threshold were observed when manipulating the characteristics of the subjects (e.g., [22]), the visual environment [26], the availability of visual feedback [27,28] or the standing postures [29]. Such a process, however, appears ineffective at limiting body sway in the Extended head position. In conclusion, in light of the CP displacements dissociation into two elementary CGv and CP − CGv motions and the recourse to parameters issued from frequency analysis and fBm modelling, the present findings evidence that the enlarged territories covered by the CP trajectories, usually observed with extended head posture, are the result of an exaggerated muscular activity and lessened capacities to detect and control the CG motions. It is concluded that the head extended posture commonly encountered in many routine activities, may represent a challenge for the postural control system, even for young healthy adults. Whether the present effects can again be observed in people showing less accurate postural capacities (e.g., elderly persons) for whom the consequences of an impaired postural control could be more dramatic remains to be investigated and is included in our immediate plans. Finally, although the extended head posture probably impacts on various postural tasks, the nature of these impacts is yet to be established. Whether and how such an extended head posture may modulate other postural control mechanisms, such as anticipatory adjustments for voluntary movements or automatic postural reactions to external perturbations, is currently being investigated.
Acknowledgements The authors would like to thank the anonymous reviewers for helpful comments and suggestions. This paper was written while the first author was ATER at Universit´e de Savoie, France.
References [1] Newell KM. Constraints on the development of coordination. In: Wade MG, Whiting HTA, editors. Motor development in children: aspects of coordination and control. Boston: Martinus Nijhoff; 1986. p. 341– 60. [2] Barin K, Seitz CM, Welling DB. Effect of head orientation on the diagnostic sensitivity of posturography in patients with compensated unilateral lesions. Otolaryngol Head Neck Surg 1992;106:355–62. [3] Brandt T, Krafczyk S, Malsbenden I. Postural imbalance with head extension: improvement by training as a model for ataxia therapy. Ann N Y Acad Sci 1981;374:636–49. [4] Jackson RT, Epstein CM. Effect of head extension on equilibrium in normal subjects. Ann Otol Rhinol Laryngol 1991;100:63–7. [5] Kogler A, Lindfors J, Odkvist LM, Ledin T. Postural stability using different neck positions in normal subjects and patients with neck trauma. Acta Otolaryngol 2000;120:151–5. [6] Norr´e ME. Head extension effect in static posturography. Ann Otol Rhinol Laryngol 1995;104:570–3.
[7] Norr´e ME, Forrez G. Posture testing (posturography) in the diagnosis of peripheral vestibular pathology. Arch Otorhinolaryngol 1986;243:186–9. [8] Simoneau GG, Leibowitz HW, Ulbrecht JS, Tyrrell RA, Cavanagh PR. The effects of visual factors and head orientation on postural steadiness in women 55 to 70 years of age. J Gerontol 1992;47:M151–8. [9] Sheldon JH. On the natural history of falls in old age. Br Med J 1960;2:1685–90. [10] Endo K, Ichimaru K, Shimura H, Imakiire A. Cervical vertigo after hair shampoo treatment at a hairdressing salon: a case report. Spine 2000;25:632–4. [11] Rougier P, Caron O. Centre of gravity motions and ankle joint stiffness control in upright undisturbed stance modelled through fractional Brownian motion framework. J Mot Behav 2000;32:405–13. [12] Winter DA, Patla AE, Prince F, Ishac M, Gielo-Perczak K. Stiffness control of balance in quiet standing. J Neurophysiol 1998;80:1211–21. [13] Cl´ement G, Gurfinkel VS, Lestienne F, Lipshits MI, Popov KE. Adaptation of postural control to weightlessness. Exp Brain Res 1984;57:61–72. [14] Horstmann A, Dietz V. A basic posture control mechanism: the stabilization of the centre of gravity. Electroenceph Clin Neurophysiol 1990;76:165–76. [15] Winter DA, Prince F, Franck JS, Powel C, Zabjek KF. Unified theory regarding A/P and M/L balance in quiet standing. J Neurophysiol 1996;75:2334–43. [16] Breni`ere Y, Do MC, Bouisset S. Are dynamic phenomena prior to stepping essential to walking? J Mot Behav 1987;19:62–76. [17] Caron O, G´elat T, Rougier P, Blanchi JP. A comparative analysis of the centre of gravity and centre of pressure sway paths in standing posture: an estimation of active stiffness. J Appl Biomech 2000;16:234–47. [18] Deligni`eres D, Deschamps T, Legros A, Caillou N. A methodological note on nonlinear time series analysis: is the open- and closedloop model of Collins and De Luca (1993) a statistical artifact? J Mot Behav 2003;35:86–97. [19] Peterka RJ. Postural control model interpretation of stabilogram diffusion analysis. Biol Cybern 2000;82:335–43. [20] Chiari L, Cappello A, Lenzi D, Della Croce U. An improved technique for the extraction of stochastic parameters from stabilograms. Gait Posture 2000;12:225–34. [21] Collins JJ, De Luca CJ. Open-loop and closed-loop control of posture: a random-walk analysis of centre-of-pressure trajectories. Exp Brain Res 1993;95:308–18. [22] Collins JJ, De Luca CJ. The effects of visual input on openloop and closed-loop postural control mechanisms. Exp Brain Res 1995;103:151–63. [23] Riley MA, Wong S, Mitra S, Turvey MT. Common effects of touch and vision on postural parameters. Exp Brain Res 1997;117:165–70. [24] Rougier P. Influence of visual feedback on successive control mechanisms in upright stance in humans assessed by fractional Brownian motion modelling. Neurosci Lett 1999;266:157–60. [25] Rougier P, Farenc I. Adaptative effects of loss of vision on upright undisturbed stance. Brain Res 2000;871:165–74. [26] Rougier P. The influence of having the eyelids open or closed on undisturbed postural control. Neurosci Res 2003;47:73–83. [27] Rougier P. Optimising the visual feedback technique for improving upright stance maintenance by delaying its display: behavioural effects on healthy adults. Gait Posture 2004;19:154–63. [28] Rougier P. Visual feedback induces opposite effects on elementary centre of gravity and centre of pressure minus centre of gravity motions in undisturbed upright stance. Clin Biomech 2003;18:341– 9. [29] Rougier P, Burdet C, Farenc I, Berger L. Backward and forward leaning postures modelled by a fBm framework. Neurosci Res 2001;41:41–50.
N. Vuillerme, P. Rougier / Gait and Posture 21 (2005) 318–325 [30] Paulus WM, Straube A, Brandt T. Visual stabilization of posture: physiological stimulus characteristics and clinical aspects. Brain 1984;107:1143–63. [31] Carpenter MG, Frank JS, Winter DA, Peysar GW. Sampling duration effects on centre of pressure summary measures. Gait Posture 2001;13:35–40. [32] Caron O, Faure B, Breni`ere Y. Estimating the center of gravity of the body on the basis of the center of pressure in standing posture. J Biomech 1997;30:1169–71. [33] Tagaki A, Fujimura E, Suehiro S. A new method of statokinesigram area measurement. Application of a statistically calculated ellipse. In: Igarashi M, Black O, editors. Vestibular and visual control on posture and locomotor equilibrium. Bˆale: Karger; 1985. p. 74–9. [34] Mandelbrot BB, van Ness JW. Fractional Brownian motions, fractional noises and applications. SIAM Rev 1968;10:422–37. [35] Ivanenko YP, Talis VL, Kazennikov OV. Support stability influences postural responses vibrations in humans. Eur J Neurosci 1999;11:647–54. [36] Roll JP, Vedel JP, Ribot E. Alteration of proprioceptive messages induced by tendon vibration in man: a microneurographic study. Environ Brain Res 1989;76:213–22. [37] Karnath HO, Sievering D, Fetter M. The interactive contribution of neck muscle proprioception and vestibular stimulation to subjective “straight ahead” orientation in man. Exp Brain Res 1994;101:140– 6.
325
[38] Strupp M, Arbusow V, Borges Pereira C, Dieterich M, Brandt T. Subjective straight-ahead during neck muscle vibration: effects of ageing. NeuroReport 1999;10:3191–4. [39] Tokizane T, Murao M, Ogata T, Kondo T. Electromyographic studies on tonic neck, lumbar, and labyrinthine reflexes in normal persons. Jpn J Physiol 1951;2:30–146. [40] Corriveau H, Prince F, Hebert R, Raiche M, Tessier D, Maheux P, et al. Evaluation of postural stability in elderly with diabetic neuropathy. Diabetes Care 2000;23:1187–91. [41] Laughton CA, Slavin M, Katdare K, Nolan L, Bean JF, Kerrigan DC, et al. Aging, muscle activity, and balance control: physiologic changes associated with balance impairment. Gait Posture 2003;180:101–8. [42] Karlberg M, Persson L, Magnusson M. Reduced postural control in patients with chronic cervicobrachial pain syndrome. Gait Posture 1995;3:241–9. [43] Michaelson P, Michaelson M, Jaric S, Latash ML, Sj¨olander P, Djupsj¨obacka M. Vertical posture and head stability in patients with chronic neck pain. J Rehabil Med 2003;35:229–35. [44] Gregoric M, Tayeka T, Baron JB, Bessineton JC. Influence of vibration of neck muscles on balance control in man. Agressologie 1978;19:37–8. [45] Koskimies K, Sutinen P, Aalto H, Starck J, Toppila E, Hirvonen T, et al. Postural stability, neck proprioception and tension neck. Acta Otolaryngol Suppl 1997;529:95–7.