- Email: [email protected]
Cognitive demands of postural control during continuous rotational perturbations of the support surface
Gait & Posture 29 (2009) 86–90
Contents lists available at ScienceDirect
Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost
Cognitive demands of postural control during continuous rotational perturbations of the support surface Sakineh B. Akram a,*, James S. Frank b a b
Department of Kinesiology, University of Waterloo, Waterloo, Ontario, Canada University of Windsor, Windsor, Ontario, Canada
A R T I C L E I N F O
A B S T R A C T
Article history: Received 22 January 2008 Received in revised form 30 June 2008 Accepted 5 July 2008
This research explored the effect of the level of balance challenge on the cognitive demands of dynamic balance. The postural task was maintaining balance while standing on a rotating platform that moved about the axis of the ankle joint in the pitch plane. Different frequencies and amplitudes of perturbation were employed to introduce different levels of balance challenge. The cognitive task was a silent word identification task. Results showed no significant difference in the participants’ performance on the cognitive task in any of the dual-task conditions in comparison with their performance in the ‘‘cognitive only’’ condition. Furthermore, regardless of the level of balance challenge concurrent performance of the cognitive task did not affect the balance control strategies adopted by our participants. The lack of interference between the cognitive task and the present postural task might be due to the fact that the rotational perturbations induced very small Center of Mass (COM) displacements at all frequencies which were under the automatic control. ß 2008 Elsevier B.V. All rights reserved.
Keywords: Cognitive Postural control Rotating platform Dual task
1. Introduction Few studies have examined the cognitive demands of postural control during dynamic balance tasks [1–3] which involve a series of ongoing perturbations such as walking. Lajoie et al. showed that in healthy young adults sitting, standing and different phases of walking represent a postural hierarchy with respect to attentional demand with the less stable postural positions requiring more attention to ensure stability [1]. We sought to explore the cognitive demands of different levels of dynamic balance challenge by requiring participants to stand on a continuously rotating platform and systematically varying the frequency and amplitude of the support surface perturbation. Walsh [4] and Diener [5] examined the control of posture on a continuously rotating platform and reported that for frequencies below 1 Hz there is a greater attenuation of body motion as the frequency increases. In an earlier study [6] we selected frequencies both below and above 1 Hz and found that balance control on a continuously rotating platform was the same across a number of frequencies and amplitudes. In general, the body Center of Mass (COM) angular displacement was attenuated across all frequencies and amplitudes of platform perturbation and was maintained close
* Corresponding author. Tel.: +519 888 4567x32601; fax: +519 885 0470. E-mail address: [email protected] (S.B. Akram). 0966-6362/$ – see front matter ß 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.gaitpost.2008.07.006
to that of quiet stance on a stationary firm surface. A multisegmental strategy involving anti-phase movements of the hip and ankle was adopted at all rotation frequencies to attenuate motion of the body COM [6]. While the control strategy was the same, the cognitive demands might be different at different frequencies and/or amplitudes of perturbation. We speculated that perturbations with higher frequency and/or amplitude should introduce greater balance challenge, requiring more attention to ensure stability. Therefore, the cognitive demand of the postural control might be greater at higher frequencies of perturbation resulting in less attenuation of the body COM motion and/or poorer performance on the cognitive task when the two tasks are performed simultaneously. Furthermore, since transition from a single-task performance to dual-task performance often occurs in everyday life, we examined the influence of accommodation to a single task (balance or cognitive) on subsequent dual-task performance. 2. Methods 2.1. Participants Ten university students, five male and five female (mean S.D. age = 20.8 2.1 years) volunteered to participate in this study. Participants were free from any neurological or musculoskeletal disorders as verified by self-report. Participants were only included if they could stand barefoot on one foot (their choice) with their eyes closed for at least 25 s, thereby ensuring a normal function of the vestibular system [7]. All procedures were approved by the Office of Research Ethics, University of Waterloo.
S.B. Akram, J.S. Frank / Gait & Posture 29 (2009) 86–90 All participants were informed about the experimental procedure before signing a consent form. 2.2. Procedure Participants were asked to stand barefoot on a servo-controlled platform capable of rotating about the pitch-axis of the ankle joints. Their feet were positioned to ensure that the ankle joint axis was aligned with the pitch axis of the platform and their arms were positioned comfortably at their sides. Participants were instructed to stand on the platform as still as possible as it rotated and to perform the cognitive task as well as they could in dual-task trials. Nothing was said or suggested to indicate the priority of the postural or the cognitive task. The cognitive task was a silent word identification task. A pre-recorded series of words were played for the participants by a portable CD player. Each series of words included a random number of words that began with the letter ‘‘s.’’ Participants were asked to silently tally the number of words beginning with the letter ‘‘s’’ and report the total count at the end of each trial. No feedback on performance was given during the testing period. The silent-counting task was chosen to eliminate any possible effect of articulation on COM position as suggested by Yardley et al. [8]. Four different combinations of frequency and amplitude of support surface rotation were employed: 0.5 Hz at 28 amplitude (amplitude = half cycle, i.e., 28 toesup and 28 toes-down), 1.0 Hz at 18, 1.5 Hz at 48, and 2 Hz at 38. Each participant was tested in the following four different task conditions performed at each of the four different levels of platform perturbation frequency/amplitude: (1) Posture Only (PO), performing only the postural task throughout the 20 s trial; (2) Simultaneous (S), performing both the postural and the cognitive tasks simultaneously throughout the 20 s trial; (3) Posture First (PF), performing the posture task for 10 s followed by the simultaneous performance of the posture and cognitive tasks for 20 s; (4) Cognitive First (CF), performing the cognitive task for 10 s followed by the simultaneous performance of the posture and cognitive tasks for 20 s. PF and CF conditions were included to examine the effect of accommodation to a single task (postural or cognitive) on subsequent dual-task performance. For these conditions, data was collected only during the concurrent performance of the cognitive and posture task (the last 20 s of the trials). Performance on the cognitive task alone was examined at the beginning of the testing session. Participants performed three consecutive 20 s trials of the cognitive task while seated (Cognitive Only or CO). Participants were tested for three trials in each task condition and at each level of perturbation frequency. Therefore, considering the three trials in the CO condition, each participant was tested for a total of 51 trials. To minimize any possible order effect, the order of the trials was completely randomized except for the cognitive only trials which were always performed at the beginning. To minimize the effect of fatigue, a 5-min rest was provided after every 10 trials. Throughout the trials, participants wore a lightweight climbing harness attached to a ceiling beam to ensure safety. 2.3. Data collection An Optotrak 3D imaging system (OPTOTRAK, Northern Digital, Waterloo, Ontario, Canada) was used to collect kinematic data at 60 Hz. One horizontal camera was placed in front of the participant to obtain a frontal view of the participant. In order to derive whole body COM, 21 infra red emitting diodes (IREDs) were placed on the 21 anatomical landmarks of participant’s body as recommended by Winter et al. [9]; see Akram et al. [6] for details of marker placements. 2.4. Data processing For each trial, angular displacement of body COM was computed assuming the position of the ankle joint was fixed on the platform. Joint angles (right ankle, knee, and hip), head and platform angular displacement also were calculated for each trial; see Akram et al. [6] for calculation details. The root mean square (RMS) of angular displacements of the COM, head, hip, knee, ankle, and platform were computed for each trial. Since perturbation amplitude differed across frequencies, the RMS values in each trial were normalized to the amplitude of platform displacement for that trial; this ratio is referred to as ‘‘RMS ratio’’ and was considered for statistical analysis. Cross correlations were computed between angular displacements of the COM and platform. Inter-joint correlations also were computed. Performance on the cognitive task was scored by absolute error, i.e., the absolute difference between the actual number of the words starting with letter ‘‘s’’ in each word series and the number which was reported by the participant. 2.5. Statistical analysis Participant’s absolute error in performing the cognitive task was averaged across the three trials performed in each task condition (CO, S, PF, CF). The resultant values were used in the Dunnett’s t-tests to compare the participants’ performance on the cognitive task in different task conditions.
87
To examine balance performance of participants in different task conditions, for each dependent variable of interest a three-way repeated measures analysis of variance (ANOVA) with gender as between factor, and task (PO, S, PF, CF) and frequency (0.5, 1, 1.5, 2 Hz) as within factors was performed. In all cases, a significance value (P) of less than 0.05 was used to test statistical significance. In conditions that a main effect of task on a particular dependent variable was revealed, Tukey’s Studentized Range (HSD) Test was performed to determine which means were significantly different from the others.
3. Results 3.1. Performance of the cognitive task Overall, the magnitude of error in performing the cognitive task was significantly higher for males than females (0.64 0.59 vs. 0.28 0.34 absolute error, F(1,8) = 10.43, P = 0.0121). However, the Dunnett’s t-tests showed no significant difference in the participants’ performance on the cognitive task in any of the dual-task conditions in comparison with their performance in cognitive only condition. Participants did not achieve perfect performance when the cognitive task was performed alone and this level of performance was not changed when the cognitive task was paired with the postural task. 3.2. Performance of the balance task Changes in the participants’ performance on the balance task in the dual-task conditions (S, PF, CF) in comparison with the PO condition at different frequencies of perturbation are reported. All significant main and interaction effects are reported here. However, since task effects are the primary interest of this paper, only the sources of the main and interaction effects of task are reported. The three-way ANOVA on the RMS ratio of body COM angular displacement showed significant effect of frequency (F(3,440) = 433.87, P < 0.0001), and task (F(3,440) = 10.66, P < 0.0001). The gender task (F(3,440) = 6.26, P = 0.0004) interaction effect was also significant. The significant gender task interaction effect was due to the smaller RMS ratio of body COM angular displacement for male participants in PF condition than in any other task condition. Mean RMS ratio of COM angular displacement for male participants was 0.30, 0.23, 0.30, and 0.32 in CF, PF, PO, and S conditions, respectively (Fig. 1a). The RMS ratios of the angular displacement of hip, knee and ankle joints were generally smallest in PF condition. The three-way ANOVA of the RMS ratio of the hip angular displacement showed significant effects of frequency (F(3,440) = 148.65, P < 0.0001), task (F(3,440) = 18.50, P < 0.0001), gender task (F(3,440) = 4.66, P = 0.0032), frequency task (F(9,440) = 3.41, P = 0.0005), and gender task frequency (F(9,440) = 2.16, P = 0.0236). Further investigation revealed that the RMS ratio of hip angular displacement was significantly smaller in PF condition than in any other task condition. Furthermore, only in PF condition the RMS ratio of hip angular displacement was the same for males and females in all frequencies of perturbation (Fig. 1b). ANOVA of the RMS ratio of angular displacement of knee joint showed a significant effect of frequency (F(3,440) = 54.33, P < 0.0001), gender frequency (F(3,440) = 3.26, P = 0.0214), and task (F(3,440) = 10.50, P < 0.0001). The significant main effect of task was due to the smaller RMS ratio of angular displacement of the knee joint in PF condition than in any other task condition. Mean RMS ratio of angular displacement of the knee joint was 0.46, 0.34, 0.49, and 0.45 in CF, PF, PO, and S conditions, respectively (Fig. 1c). ANOVA of the RMS ratio of angular displacement of the ankle joint revealed a significant effect of frequency (F(3,440) = 126.63, P < 0.0001), gender frequency (F(3,440) = 7.20, P = 0.0001), and
88
S.B. Akram, J.S. Frank / Gait & Posture 29 (2009) 86–90
Fig. 1. Mean and standard error of the mean for RMS ratio of angular displacement of body COM (a), hip (b), knee (c), and ankle (d) for male and female participants at four different frequencies of platform perturbation across four different task conditions. (CF: Cognitive First; PF: Posture First; PO: Posture Only; S: Simultaneous).
task (F(3,440) = 4.03, P = 0.0076). The significant main effect of task was due to the smaller RMS ratio of angular displacement of the ankle joint in PF condition than in PO and CF conditions. RMS ratio of angular displacement of the ankle joint in S condition was not different from the comparable value in any other task
condition (Fig. 1d). Mean RMS ratio of angular displacement of the ankle joint was 0.85, 0.80, 0.85, and 0.83, in CF, PF, PO, and S conditions. Task had no main or interaction effect on RMS ratio of head angular displacement in pitch plane.
S.B. Akram, J.S. Frank / Gait & Posture 29 (2009) 86–90
Correlations between angular displacement of body COM and the platform were poor (ranging from 0.06 to 0.09) and remained unchanged across all task conditions. The highest inter-joint correlations were between ankle and hip, and did not differ across tasks (mean = 0.33). 4. Discussion The objective of the present study was to examine the effect of the level of balance challenge on the cognitive demands of dynamic balance by systematically varying the frequency and amplitude of support surface perturbation. The effect of accommodation to a single task (postural or cognitive) on subsequent dual-task performance was also examined. The main findings were: (1) Regardless of the frequency and amplitude of perturbation, the introduction of a cognitive task did not modify the balance control strategy during simultaneous performance of the tasks or when the postural task was introduced after initiating the cognitive task. (2) PF condition was performed differently than other task conditions. Regardless of the task condition, there was a significant increase in the RMS ratios of body COM, hip, knee, and ankle angular displacements during 1 Hz perturbations. This phenomenon has been noted for quiet stance on an unstable support surface [10,11]. Creath et al. reported that 1 Hz frequency represents a changeover point between the use of an ankle to an ankle–hip strategy [12]. Explanations for the greater displacement at 1 Hz range from changes in sensory feedback gains controlling balance [10], to a natural frequency of the body biomechanics [11], or a combination of the two [12]. This effect has been reported in our previous paper [6]. Participants’ performance on the balance task was compared among different task conditions (PO, S, PF, and CF). In the Posture Only condition, participants adopted a multi-segmental strategy involving anti-phase movements of the hip and ankle joints to attenuate motion of the body COM. The body COM angular displacement was attenuated across all frequencies of platform perturbation and was maintained close to that of quiet stance on a stationary firm surface [6]. The introduction of a cognitive task simultaneous or in advance of platform movement had no effect on COM displacement or postural control strategy as revealed by joint angular displacements. Participants’ balance performance in PO, S, and CF conditions was not different from each other, suggesting that addition of the cognitive task had no significant effect on postural balance. Studies on the effects of simultaneous performance of a cognitive task during quiet stance showed changes in the performance of either the balance task [13], or the cognitive task [1], or both tasks [14]. However, results of the present study did not show any interference between the cognitive and postural tasks while participants were standing on a rotating platform. Cognitive performance remained the same across different combinations of amplitude/frequency of support surface perturbations. Furthermore, the introduction of the cognitive task (simultaneous or in advance of platform movement) did not change balance performance of participants. It should be noted that none of the participants had a perfect performance in the CO condition in which they performed only the cognitive task while seated, suggesting that the cognitive task was difficult. The lack of interference between the cognitive task and the present postural task might be due to the fact that the rotational perturbations induced very small COM displacement, particularly at higher perturbation frequencies. Gage et al. [15] reported that during quiet stance, the mean RMS of body COM is 6 mm or 0.388 (assuming COM height of 90 cm relative to the ankle joints). In our
89
study, averaged across all task conditions, the mean RMS of body COM angular displacement was 0.568, 0.498, 0.548 and 0.518 for 0.5, 1, 1.5 and 2 Hz perturbations, respectively. It is possible that the perturbation frequencies employed in the present study did not impose different levels of balance challenge. The findings of the present study may also be related to the participants’ focus of attention. During quiet stance, participants try to control the destabilizing effects of their own movement to maintain their balance, while when faced with platform perturbations participants try to control and minimize the effects of an external destabilizing force on their postural balance. Unlike focusing on one’s own movement as occurs during quiet standing (internal focus), focusing on the effect of an external perturbation (external focus) may result in ‘‘more effective and natural interplay between voluntary and reflexive control processes’’ [16] which allow balance control with minimum cognitive requirements. As a result of this greater degree of postural control automaticity, the addition of a cognitive task does not degrade performance of either the cognitive or the balance task [16]. The unexpected introduction of a competing cognitive task (as in PF condition) resulted in smaller joint displacements for both males and females and, smaller COM displacement (tighter control) for males. The RMS ratio of body COM angular displacement remained the same across different task conditions for females; however, the male participants showed significantly larger body movements in all task conditions except the PF condition. Males were taller and heavier than females (mean S.D. height = 182.5 6.3 and 167.1 9.4 cm; mean S.D. weight = 75 8.3 and 60.9 9.9 kg for males and females, respectively) which could explain the larger COM displacement in the S, CF, and PO condition. For male participants the smaller RMS ratio of body COM angular displacement in PF condition is explained by displacement at the knee joint. As shown in Fig. 1c, for both males and females the RMS ratio of angular displacement of the knee joint was smaller in PF condition than in any other task condition; however, the magnitude of change in the RMS ratio of angular displacement of the knee joint was much greater for males than females and large enough to significantly reduce the movement of the body COM. Nevertheless, the gender effect should be interpreted with caution because of the small number of participants in each gender group. Although the variability of RMS ratio of COM and lower limb joints differed across different task conditions, these changes were not significant enough to elicit alternative balance strategy. In all different task conditions movements of the hip and ankle were adopted to attenuate the movement of the body COM. The phase pattern between movements of the hip and ankle was similar across all task conditions (Fig. 2). Task had no main or interaction effect on RMS ratio of head angular displacement in pitch plane. This finding could indicate that participants kept their head stable in different task conditions to minimize the disturbing effects of head movements on the two main sensory systems nested in the head (vision and vestibular systems). In summary, the present study showed that the addition of a concurrent cognitive task does not affect the balance strategies adopted by young healthy adults while challenged by continuous rotational perturbations of the support surface. Systematically varying the frequency and amplitude of support surface perturbation does not elicit interference between the two tasks. This lack of interference may reflect more automatic control of this balance task. This study also showed that while accommodation to a cognitive task does not have any effect on subsequent dual-task performance, accommodation to a balance task results in a tighter control of body during the subsequent dual-task performance.
90
S.B. Akram, J.S. Frank / Gait & Posture 29 (2009) 86–90
Fig. 2. Sample record of angular displacements of ankle and hip joints of a representative participant during four trials performed at the four different task conditions (CF: Cognitive First; PF: Posture First; PO: Posture Only; S: Simultaneous) during 1.5 Hz platform perturbation.
Acknowledgement The authors would like to acknowledge the Natural Sciences and Engineering Research Council (NSERC) for the funding of this research. Conflict of interest The authors declare that they have no conflict of interest, financial or otherwise, related to the submitted manuscript or the associated research. References [1] Lajoie Y, Teasdale N, Bard C, Fleury M. Attentional demands for static and dynamic equilibrium. Exp Brain Res 1993;97:139–44. [2] Chen HC, Schultz AB, Ashton-Miller JA, Giordani B, Alexander NB, Guire KE. Stepping over obstacles: dividing attention impairs performance of old more than young adults. J Gerontol A Biol Sci Med Sci 1996;51:M116–22. [3] Gage WH, Sleik RJ, Polych MA, McKenzie NC, Brown LA. The allocation of attention during locomotion is altered by anxiety. Exp Brain Res 2003;150: 385–94. [4] Walsh EG. Standing man, slow rhythmic tilting, importance of vision. Agressologie 1973;14C:79–85.
[5] Diener HC, Dichgans J, Bruzek W, Selinka H. Stabilization of human posture during induced oscillations of the body. Exp Brain Res 1982;45:126–32. [6] Akram SB, Frank JS, Patla AE, Allum JH. Balance control during continuous rotational perturbations of the support surface. Gait Posture 2008;27:393–8. [7] Allum JH, Adkin AL, Carpenter MG, Held-Ziolkowska M, Honegger F, Pierchala K. Trunk sway measures of postural stability during clinical balance tests: effects of a unilateral vestibular deficit. Gait Posture 2001;14:227–37. [8] Yardley L, Gardner M, Leadbetter A, Lavie N. Effect of articulatory and mental tasks on postural control. Neuroreport 1999;10:215–9. [9] Winter DA, Patla AE, Prince F, Ishac M, Gielo-Perczak K. Stiffness control of balance in quiet standing. J Neurophysiol 1998;80:1211–21. [10] Peterka RJ, Loughlin PJ. Dynamic regulation of sensori-motor integration in human postural control. J Neurophysiol 2004;91:410–23. [11] Allum JHJ, Zamani F, Adkin AL, Ernst A. Differences between trunk sway characteristics on a foam support surface and on the Equitest1 anklesway-referenced support surface. Gait Posture 2002;16:264–70. [12] Creath R, Kiemal T, Horak F, Peterka R, Jeka J. A unified view of quiet and perturbed stance: simultaneous co-existing excitable modes. Neurosci Lett 2005;377:75–80. [13] Shumway-Cook A, Woollacott M, Kerns KA, Baldwin M. The effects of two types of cognitive tasks on postural stability in older adults with and without a history of falls. J Gerontol A Biol Sci Med Sci 1997;52:M232–40. [14] Brauer SG, Woollacott M, Shumway-Cook A. The interacting effects of cognitive demand and recovery of postural stability in balance-impaired elderly persons. J Gerontol A Biol Sci Med Sci 2001;56:M489–96. [15] Gage WH, Winter DA, Frank JS, Adkin AL. Kinematic and kinetic validity of the inverted pendulum model in quiet standing. Gait Posture 2004;19:124–32. [16] Wulf G, McNevin N, Shea CH. The automaticity of complex motor skill learning as a function of attentional focus. Q J Exp Psychol A 2001;54:1143–54.
Contents lists available at ScienceDirect
Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost
Cognitive demands of postural control during continuous rotational perturbations of the support surface Sakineh B. Akram a,*, James S. Frank b a b
Department of Kinesiology, University of Waterloo, Waterloo, Ontario, Canada University of Windsor, Windsor, Ontario, Canada
A R T I C L E I N F O
A B S T R A C T
Article history: Received 22 January 2008 Received in revised form 30 June 2008 Accepted 5 July 2008
This research explored the effect of the level of balance challenge on the cognitive demands of dynamic balance. The postural task was maintaining balance while standing on a rotating platform that moved about the axis of the ankle joint in the pitch plane. Different frequencies and amplitudes of perturbation were employed to introduce different levels of balance challenge. The cognitive task was a silent word identification task. Results showed no significant difference in the participants’ performance on the cognitive task in any of the dual-task conditions in comparison with their performance in the ‘‘cognitive only’’ condition. Furthermore, regardless of the level of balance challenge concurrent performance of the cognitive task did not affect the balance control strategies adopted by our participants. The lack of interference between the cognitive task and the present postural task might be due to the fact that the rotational perturbations induced very small Center of Mass (COM) displacements at all frequencies which were under the automatic control. ß 2008 Elsevier B.V. All rights reserved.
Keywords: Cognitive Postural control Rotating platform Dual task
1. Introduction Few studies have examined the cognitive demands of postural control during dynamic balance tasks [1–3] which involve a series of ongoing perturbations such as walking. Lajoie et al. showed that in healthy young adults sitting, standing and different phases of walking represent a postural hierarchy with respect to attentional demand with the less stable postural positions requiring more attention to ensure stability [1]. We sought to explore the cognitive demands of different levels of dynamic balance challenge by requiring participants to stand on a continuously rotating platform and systematically varying the frequency and amplitude of the support surface perturbation. Walsh [4] and Diener [5] examined the control of posture on a continuously rotating platform and reported that for frequencies below 1 Hz there is a greater attenuation of body motion as the frequency increases. In an earlier study [6] we selected frequencies both below and above 1 Hz and found that balance control on a continuously rotating platform was the same across a number of frequencies and amplitudes. In general, the body Center of Mass (COM) angular displacement was attenuated across all frequencies and amplitudes of platform perturbation and was maintained close
* Corresponding author. Tel.: +519 888 4567x32601; fax: +519 885 0470. E-mail address: [email protected] (S.B. Akram). 0966-6362/$ – see front matter ß 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.gaitpost.2008.07.006
to that of quiet stance on a stationary firm surface. A multisegmental strategy involving anti-phase movements of the hip and ankle was adopted at all rotation frequencies to attenuate motion of the body COM [6]. While the control strategy was the same, the cognitive demands might be different at different frequencies and/or amplitudes of perturbation. We speculated that perturbations with higher frequency and/or amplitude should introduce greater balance challenge, requiring more attention to ensure stability. Therefore, the cognitive demand of the postural control might be greater at higher frequencies of perturbation resulting in less attenuation of the body COM motion and/or poorer performance on the cognitive task when the two tasks are performed simultaneously. Furthermore, since transition from a single-task performance to dual-task performance often occurs in everyday life, we examined the influence of accommodation to a single task (balance or cognitive) on subsequent dual-task performance. 2. Methods 2.1. Participants Ten university students, five male and five female (mean S.D. age = 20.8 2.1 years) volunteered to participate in this study. Participants were free from any neurological or musculoskeletal disorders as verified by self-report. Participants were only included if they could stand barefoot on one foot (their choice) with their eyes closed for at least 25 s, thereby ensuring a normal function of the vestibular system [7]. All procedures were approved by the Office of Research Ethics, University of Waterloo.
S.B. Akram, J.S. Frank / Gait & Posture 29 (2009) 86–90 All participants were informed about the experimental procedure before signing a consent form. 2.2. Procedure Participants were asked to stand barefoot on a servo-controlled platform capable of rotating about the pitch-axis of the ankle joints. Their feet were positioned to ensure that the ankle joint axis was aligned with the pitch axis of the platform and their arms were positioned comfortably at their sides. Participants were instructed to stand on the platform as still as possible as it rotated and to perform the cognitive task as well as they could in dual-task trials. Nothing was said or suggested to indicate the priority of the postural or the cognitive task. The cognitive task was a silent word identification task. A pre-recorded series of words were played for the participants by a portable CD player. Each series of words included a random number of words that began with the letter ‘‘s.’’ Participants were asked to silently tally the number of words beginning with the letter ‘‘s’’ and report the total count at the end of each trial. No feedback on performance was given during the testing period. The silent-counting task was chosen to eliminate any possible effect of articulation on COM position as suggested by Yardley et al. [8]. Four different combinations of frequency and amplitude of support surface rotation were employed: 0.5 Hz at 28 amplitude (amplitude = half cycle, i.e., 28 toesup and 28 toes-down), 1.0 Hz at 18, 1.5 Hz at 48, and 2 Hz at 38. Each participant was tested in the following four different task conditions performed at each of the four different levels of platform perturbation frequency/amplitude: (1) Posture Only (PO), performing only the postural task throughout the 20 s trial; (2) Simultaneous (S), performing both the postural and the cognitive tasks simultaneously throughout the 20 s trial; (3) Posture First (PF), performing the posture task for 10 s followed by the simultaneous performance of the posture and cognitive tasks for 20 s; (4) Cognitive First (CF), performing the cognitive task for 10 s followed by the simultaneous performance of the posture and cognitive tasks for 20 s. PF and CF conditions were included to examine the effect of accommodation to a single task (postural or cognitive) on subsequent dual-task performance. For these conditions, data was collected only during the concurrent performance of the cognitive and posture task (the last 20 s of the trials). Performance on the cognitive task alone was examined at the beginning of the testing session. Participants performed three consecutive 20 s trials of the cognitive task while seated (Cognitive Only or CO). Participants were tested for three trials in each task condition and at each level of perturbation frequency. Therefore, considering the three trials in the CO condition, each participant was tested for a total of 51 trials. To minimize any possible order effect, the order of the trials was completely randomized except for the cognitive only trials which were always performed at the beginning. To minimize the effect of fatigue, a 5-min rest was provided after every 10 trials. Throughout the trials, participants wore a lightweight climbing harness attached to a ceiling beam to ensure safety. 2.3. Data collection An Optotrak 3D imaging system (OPTOTRAK, Northern Digital, Waterloo, Ontario, Canada) was used to collect kinematic data at 60 Hz. One horizontal camera was placed in front of the participant to obtain a frontal view of the participant. In order to derive whole body COM, 21 infra red emitting diodes (IREDs) were placed on the 21 anatomical landmarks of participant’s body as recommended by Winter et al. [9]; see Akram et al. [6] for details of marker placements. 2.4. Data processing For each trial, angular displacement of body COM was computed assuming the position of the ankle joint was fixed on the platform. Joint angles (right ankle, knee, and hip), head and platform angular displacement also were calculated for each trial; see Akram et al. [6] for calculation details. The root mean square (RMS) of angular displacements of the COM, head, hip, knee, ankle, and platform were computed for each trial. Since perturbation amplitude differed across frequencies, the RMS values in each trial were normalized to the amplitude of platform displacement for that trial; this ratio is referred to as ‘‘RMS ratio’’ and was considered for statistical analysis. Cross correlations were computed between angular displacements of the COM and platform. Inter-joint correlations also were computed. Performance on the cognitive task was scored by absolute error, i.e., the absolute difference between the actual number of the words starting with letter ‘‘s’’ in each word series and the number which was reported by the participant. 2.5. Statistical analysis Participant’s absolute error in performing the cognitive task was averaged across the three trials performed in each task condition (CO, S, PF, CF). The resultant values were used in the Dunnett’s t-tests to compare the participants’ performance on the cognitive task in different task conditions.
87
To examine balance performance of participants in different task conditions, for each dependent variable of interest a three-way repeated measures analysis of variance (ANOVA) with gender as between factor, and task (PO, S, PF, CF) and frequency (0.5, 1, 1.5, 2 Hz) as within factors was performed. In all cases, a significance value (P) of less than 0.05 was used to test statistical significance. In conditions that a main effect of task on a particular dependent variable was revealed, Tukey’s Studentized Range (HSD) Test was performed to determine which means were significantly different from the others.
3. Results 3.1. Performance of the cognitive task Overall, the magnitude of error in performing the cognitive task was significantly higher for males than females (0.64 0.59 vs. 0.28 0.34 absolute error, F(1,8) = 10.43, P = 0.0121). However, the Dunnett’s t-tests showed no significant difference in the participants’ performance on the cognitive task in any of the dual-task conditions in comparison with their performance in cognitive only condition. Participants did not achieve perfect performance when the cognitive task was performed alone and this level of performance was not changed when the cognitive task was paired with the postural task. 3.2. Performance of the balance task Changes in the participants’ performance on the balance task in the dual-task conditions (S, PF, CF) in comparison with the PO condition at different frequencies of perturbation are reported. All significant main and interaction effects are reported here. However, since task effects are the primary interest of this paper, only the sources of the main and interaction effects of task are reported. The three-way ANOVA on the RMS ratio of body COM angular displacement showed significant effect of frequency (F(3,440) = 433.87, P < 0.0001), and task (F(3,440) = 10.66, P < 0.0001). The gender task (F(3,440) = 6.26, P = 0.0004) interaction effect was also significant. The significant gender task interaction effect was due to the smaller RMS ratio of body COM angular displacement for male participants in PF condition than in any other task condition. Mean RMS ratio of COM angular displacement for male participants was 0.30, 0.23, 0.30, and 0.32 in CF, PF, PO, and S conditions, respectively (Fig. 1a). The RMS ratios of the angular displacement of hip, knee and ankle joints were generally smallest in PF condition. The three-way ANOVA of the RMS ratio of the hip angular displacement showed significant effects of frequency (F(3,440) = 148.65, P < 0.0001), task (F(3,440) = 18.50, P < 0.0001), gender task (F(3,440) = 4.66, P = 0.0032), frequency task (F(9,440) = 3.41, P = 0.0005), and gender task frequency (F(9,440) = 2.16, P = 0.0236). Further investigation revealed that the RMS ratio of hip angular displacement was significantly smaller in PF condition than in any other task condition. Furthermore, only in PF condition the RMS ratio of hip angular displacement was the same for males and females in all frequencies of perturbation (Fig. 1b). ANOVA of the RMS ratio of angular displacement of knee joint showed a significant effect of frequency (F(3,440) = 54.33, P < 0.0001), gender frequency (F(3,440) = 3.26, P = 0.0214), and task (F(3,440) = 10.50, P < 0.0001). The significant main effect of task was due to the smaller RMS ratio of angular displacement of the knee joint in PF condition than in any other task condition. Mean RMS ratio of angular displacement of the knee joint was 0.46, 0.34, 0.49, and 0.45 in CF, PF, PO, and S conditions, respectively (Fig. 1c). ANOVA of the RMS ratio of angular displacement of the ankle joint revealed a significant effect of frequency (F(3,440) = 126.63, P < 0.0001), gender frequency (F(3,440) = 7.20, P = 0.0001), and
88
S.B. Akram, J.S. Frank / Gait & Posture 29 (2009) 86–90
Fig. 1. Mean and standard error of the mean for RMS ratio of angular displacement of body COM (a), hip (b), knee (c), and ankle (d) for male and female participants at four different frequencies of platform perturbation across four different task conditions. (CF: Cognitive First; PF: Posture First; PO: Posture Only; S: Simultaneous).
task (F(3,440) = 4.03, P = 0.0076). The significant main effect of task was due to the smaller RMS ratio of angular displacement of the ankle joint in PF condition than in PO and CF conditions. RMS ratio of angular displacement of the ankle joint in S condition was not different from the comparable value in any other task
condition (Fig. 1d). Mean RMS ratio of angular displacement of the ankle joint was 0.85, 0.80, 0.85, and 0.83, in CF, PF, PO, and S conditions. Task had no main or interaction effect on RMS ratio of head angular displacement in pitch plane.
S.B. Akram, J.S. Frank / Gait & Posture 29 (2009) 86–90
Correlations between angular displacement of body COM and the platform were poor (ranging from 0.06 to 0.09) and remained unchanged across all task conditions. The highest inter-joint correlations were between ankle and hip, and did not differ across tasks (mean = 0.33). 4. Discussion The objective of the present study was to examine the effect of the level of balance challenge on the cognitive demands of dynamic balance by systematically varying the frequency and amplitude of support surface perturbation. The effect of accommodation to a single task (postural or cognitive) on subsequent dual-task performance was also examined. The main findings were: (1) Regardless of the frequency and amplitude of perturbation, the introduction of a cognitive task did not modify the balance control strategy during simultaneous performance of the tasks or when the postural task was introduced after initiating the cognitive task. (2) PF condition was performed differently than other task conditions. Regardless of the task condition, there was a significant increase in the RMS ratios of body COM, hip, knee, and ankle angular displacements during 1 Hz perturbations. This phenomenon has been noted for quiet stance on an unstable support surface [10,11]. Creath et al. reported that 1 Hz frequency represents a changeover point between the use of an ankle to an ankle–hip strategy [12]. Explanations for the greater displacement at 1 Hz range from changes in sensory feedback gains controlling balance [10], to a natural frequency of the body biomechanics [11], or a combination of the two [12]. This effect has been reported in our previous paper [6]. Participants’ performance on the balance task was compared among different task conditions (PO, S, PF, and CF). In the Posture Only condition, participants adopted a multi-segmental strategy involving anti-phase movements of the hip and ankle joints to attenuate motion of the body COM. The body COM angular displacement was attenuated across all frequencies of platform perturbation and was maintained close to that of quiet stance on a stationary firm surface [6]. The introduction of a cognitive task simultaneous or in advance of platform movement had no effect on COM displacement or postural control strategy as revealed by joint angular displacements. Participants’ balance performance in PO, S, and CF conditions was not different from each other, suggesting that addition of the cognitive task had no significant effect on postural balance. Studies on the effects of simultaneous performance of a cognitive task during quiet stance showed changes in the performance of either the balance task [13], or the cognitive task [1], or both tasks [14]. However, results of the present study did not show any interference between the cognitive and postural tasks while participants were standing on a rotating platform. Cognitive performance remained the same across different combinations of amplitude/frequency of support surface perturbations. Furthermore, the introduction of the cognitive task (simultaneous or in advance of platform movement) did not change balance performance of participants. It should be noted that none of the participants had a perfect performance in the CO condition in which they performed only the cognitive task while seated, suggesting that the cognitive task was difficult. The lack of interference between the cognitive task and the present postural task might be due to the fact that the rotational perturbations induced very small COM displacement, particularly at higher perturbation frequencies. Gage et al. [15] reported that during quiet stance, the mean RMS of body COM is 6 mm or 0.388 (assuming COM height of 90 cm relative to the ankle joints). In our
89
study, averaged across all task conditions, the mean RMS of body COM angular displacement was 0.568, 0.498, 0.548 and 0.518 for 0.5, 1, 1.5 and 2 Hz perturbations, respectively. It is possible that the perturbation frequencies employed in the present study did not impose different levels of balance challenge. The findings of the present study may also be related to the participants’ focus of attention. During quiet stance, participants try to control the destabilizing effects of their own movement to maintain their balance, while when faced with platform perturbations participants try to control and minimize the effects of an external destabilizing force on their postural balance. Unlike focusing on one’s own movement as occurs during quiet standing (internal focus), focusing on the effect of an external perturbation (external focus) may result in ‘‘more effective and natural interplay between voluntary and reflexive control processes’’ [16] which allow balance control with minimum cognitive requirements. As a result of this greater degree of postural control automaticity, the addition of a cognitive task does not degrade performance of either the cognitive or the balance task [16]. The unexpected introduction of a competing cognitive task (as in PF condition) resulted in smaller joint displacements for both males and females and, smaller COM displacement (tighter control) for males. The RMS ratio of body COM angular displacement remained the same across different task conditions for females; however, the male participants showed significantly larger body movements in all task conditions except the PF condition. Males were taller and heavier than females (mean S.D. height = 182.5 6.3 and 167.1 9.4 cm; mean S.D. weight = 75 8.3 and 60.9 9.9 kg for males and females, respectively) which could explain the larger COM displacement in the S, CF, and PO condition. For male participants the smaller RMS ratio of body COM angular displacement in PF condition is explained by displacement at the knee joint. As shown in Fig. 1c, for both males and females the RMS ratio of angular displacement of the knee joint was smaller in PF condition than in any other task condition; however, the magnitude of change in the RMS ratio of angular displacement of the knee joint was much greater for males than females and large enough to significantly reduce the movement of the body COM. Nevertheless, the gender effect should be interpreted with caution because of the small number of participants in each gender group. Although the variability of RMS ratio of COM and lower limb joints differed across different task conditions, these changes were not significant enough to elicit alternative balance strategy. In all different task conditions movements of the hip and ankle were adopted to attenuate the movement of the body COM. The phase pattern between movements of the hip and ankle was similar across all task conditions (Fig. 2). Task had no main or interaction effect on RMS ratio of head angular displacement in pitch plane. This finding could indicate that participants kept their head stable in different task conditions to minimize the disturbing effects of head movements on the two main sensory systems nested in the head (vision and vestibular systems). In summary, the present study showed that the addition of a concurrent cognitive task does not affect the balance strategies adopted by young healthy adults while challenged by continuous rotational perturbations of the support surface. Systematically varying the frequency and amplitude of support surface perturbation does not elicit interference between the two tasks. This lack of interference may reflect more automatic control of this balance task. This study also showed that while accommodation to a cognitive task does not have any effect on subsequent dual-task performance, accommodation to a balance task results in a tighter control of body during the subsequent dual-task performance.
90
S.B. Akram, J.S. Frank / Gait & Posture 29 (2009) 86–90
Fig. 2. Sample record of angular displacements of ankle and hip joints of a representative participant during four trials performed at the four different task conditions (CF: Cognitive First; PF: Posture First; PO: Posture Only; S: Simultaneous) during 1.5 Hz platform perturbation.
Acknowledgement The authors would like to acknowledge the Natural Sciences and Engineering Research Council (NSERC) for the funding of this research. Conflict of interest The authors declare that they have no conflict of interest, financial or otherwise, related to the submitted manuscript or the associated research. References [1] Lajoie Y, Teasdale N, Bard C, Fleury M. Attentional demands for static and dynamic equilibrium. Exp Brain Res 1993;97:139–44. [2] Chen HC, Schultz AB, Ashton-Miller JA, Giordani B, Alexander NB, Guire KE. Stepping over obstacles: dividing attention impairs performance of old more than young adults. J Gerontol A Biol Sci Med Sci 1996;51:M116–22. [3] Gage WH, Sleik RJ, Polych MA, McKenzie NC, Brown LA. The allocation of attention during locomotion is altered by anxiety. Exp Brain Res 2003;150: 385–94. [4] Walsh EG. Standing man, slow rhythmic tilting, importance of vision. Agressologie 1973;14C:79–85.
[5] Diener HC, Dichgans J, Bruzek W, Selinka H. Stabilization of human posture during induced oscillations of the body. Exp Brain Res 1982;45:126–32. [6] Akram SB, Frank JS, Patla AE, Allum JH. Balance control during continuous rotational perturbations of the support surface. Gait Posture 2008;27:393–8. [7] Allum JH, Adkin AL, Carpenter MG, Held-Ziolkowska M, Honegger F, Pierchala K. Trunk sway measures of postural stability during clinical balance tests: effects of a unilateral vestibular deficit. Gait Posture 2001;14:227–37. [8] Yardley L, Gardner M, Leadbetter A, Lavie N. Effect of articulatory and mental tasks on postural control. Neuroreport 1999;10:215–9. [9] Winter DA, Patla AE, Prince F, Ishac M, Gielo-Perczak K. Stiffness control of balance in quiet standing. J Neurophysiol 1998;80:1211–21. [10] Peterka RJ, Loughlin PJ. Dynamic regulation of sensori-motor integration in human postural control. J Neurophysiol 2004;91:410–23. [11] Allum JHJ, Zamani F, Adkin AL, Ernst A. Differences between trunk sway characteristics on a foam support surface and on the Equitest1 anklesway-referenced support surface. Gait Posture 2002;16:264–70. [12] Creath R, Kiemal T, Horak F, Peterka R, Jeka J. A unified view of quiet and perturbed stance: simultaneous co-existing excitable modes. Neurosci Lett 2005;377:75–80. [13] Shumway-Cook A, Woollacott M, Kerns KA, Baldwin M. The effects of two types of cognitive tasks on postural stability in older adults with and without a history of falls. J Gerontol A Biol Sci Med Sci 1997;52:M232–40. [14] Brauer SG, Woollacott M, Shumway-Cook A. The interacting effects of cognitive demand and recovery of postural stability in balance-impaired elderly persons. J Gerontol A Biol Sci Med Sci 2001;56:M489–96. [15] Gage WH, Winter DA, Frank JS, Adkin AL. Kinematic and kinetic validity of the inverted pendulum model in quiet standing. Gait Posture 2004;19:124–32. [16] Wulf G, McNevin N, Shea CH. The automaticity of complex motor skill learning as a function of attentional focus. Q J Exp Psychol A 2001;54:1143–54.