- Email: [email protected]
The effect of changed visual feedback on intention tremor in multiple sclerosis
Neuroscience Letters 394 (2006) 17–21
The effect of changed visual feedback on intention tremor in multiple sclerosis Peter Feys a,c,∗ , Werner Helsen a , Martinus Buekers a , Tanja Ceux a , Elke Heremans a , Bart Nuttin b , Pierre Ketelaer c , Xuguang Liu d a
Katholieke Universiteit Leuven, Department of Biomedical Kinesiology, Motor Learning Laboratory, Tervuursevest 101, 3001 Leuven, Belgium b Katholieke Universiteit Leuven, Department of Neurosciences, Belgium c National Multiple Sclerosis Center, Melsbroek, Belgium d Imperial College, Department of Neurosciences, Charing Cross Hospital, UK Received 8 July 2005; received in revised form 7 September 2005; accepted 2 October 2005
Abstract In patients with multiple sclerosis (MS), intention tremor amplitude is enhanced during the visually guided compared to the memory guided motor tasks. In the present study, the effect of visual feedback on intention tremor was investigated during visually guided wrist step-tracking tasks. Specifically, visual feedback of the hand was provided either instantly or averaged over different time windows. Thirteen MS patients with intention tremor and 14 healthy controls performed the wrist step-tracking task, while the visual representation of the actual hand position was displayed instantly or averaged over time windows of 150, 250 and 350 ms. It has been found in the patient group that, in association with a decreased initial error and decreased tremor amplitude on the screen, the amplitude of the actual performed tremor also decreased when visual feedback was changed. The tremor reduction was not different between conditions with manipulated feedback, although delays in presenting visual feedback of the hand position increased when the time window was larger. The reduction in overall tremor amplitude was unlikely related to other factors, such as eye fixation deficits or the speed of the primary hand movement. These results suggest that hand tremor severity is dependent on the visual feedback of position and movement errors. © 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Intention tremor; Visual feedback; Multiple sclerosis; Step-tracking; Delay
Upper limb intention tremor, clinically defined as an increase in tremor amplitude towards the termination of a visually guided, goal-directed movement, is frequently encountered in multiple sclerosis (MS) [1,21]. This action tremor interferes with the performance of many daily life activities, such as drinking or mouse-driven PC interaction [10,11]. On neurophysiological level, intention tremor is related to the dysfunctioning of the cerebellum and/or afferent and efferent pathways [1,9]. The cerebellar system is heavily involved in the online control of visually guided, goal-directed movements, as the cerebellum is regarded as the neural locus of internal models that mutually connect sensory information and motor actions [3,13,17,25]. Previous research has shown that MS intention tremor is greatly enhanced during the visually guided in comparison to
∗
Corresponding author. Tel.: +32 16 329098. E-mail address: [email protected] (P. Feys).
0304-3940/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2005.10.010
the memory guided, goal-directed movements [6,15,20]. This robust finding was observed after complete occlusion of the target or movement cue during both continuous ramp-tracking with a moving target [15] and discrete step-tracking of stationary targets [6]. Also a partial occlusion of the movement cue, in the spatial area between or around the target positions, induced a reduction of tremor during step-tracking [6]. In healthy subjects, visual feedback control of slow ramp-tracking movements is dependent upon visually comparing both the target and limb positions to detect errors in the ongoing movement [22]. The tracking accuracy is then obtained by voluntary error correction movements. In MS patients, intention tremor is closely related to the execution of voluntary goal-directed movements [1]. The exaggeration of overall tremor amplitude during visually guided movements was caused likely by visually dependent voluntary error correction movements [6]. The present study extends the previous research on the influence of visual feedback on intention tremor in MS by changing the visual feedback
18
P. Feys et al. / Neuroscience Letters 394 (2006) 17–21
of the hand movement. The visual dependency of tremor amplitude on feedback of the ongoing movement is investigated by manipulating the visual representation of the hand movement during a wrist step-tracking task [6,7]. In addition to the hand movements, eye movements are also recorded as increased postsaccadic unsteady fixation was shown to influence the overall tremor amplitude in MS patients [7]. Eye and hand movements were simultaneously recorded using a combined eye-head tracker (66 Hz) and an angle encoder (200 Hz), respectively. The angle encoder was mounted on a wrist orthosis that allowed flexion and extension movements. The hand position was represented on a computer screen by an unfilled circle, which allowed us to apply dynamic changes to the visual feedback. Two stationary targets were displayed at the left and right side of the screen with an intertarget distance of 200 mm, corresponding to 40◦ wrist angle. The subjects were instructed to place the circle as accurately as possible at the target and to initiate the movement to the alternate target as soon as a beep was provided by a metronome (0.25 Hz or 2 s between two consecutive beeps). A wooden cover prevented the participant from direct sight of the actual wrist movements. Besides an instant visual display of the hand position, the visual feedback of the actual hand position was manipulated by representing the average of the tracking movement over preselected time windows. Specifically, the hand positions were averaged over windows of 150, 250, and 350 ms. The time windows were chosen based on the typical frequency range of intention tremor (3–6 Hz) [4] to cover one or more tremor cycles. As such, the averaging procedure visually “damped” the amplitude of the low-frequency tremor; however, also represented it with some time delay. For instance, if a patient’s tremor has a frequency of 4 Hz, one tremor cycle has a duration of 250 ms. In this case, tremor amplitude will be visually smaller when averaged over a time window of 250 ms, and be represented with a delay that is dependent on the movement characteristics, such as speed and amplitude (see Fig. 1 for illustration). Taken together, comparisons of the wrist step-tracking performance were made across
Fig. 1. Illustration of the actual hand movement (1 ms, black) of a patient case with marked tremor and its representation on the visual display when averaged over a time window of 150 ms (green), 250 ms (blue) and 350 ms (red).
four visual conditions: one without (time window of 1 ms) and three conditions with manipulation of the visual feedback (time windows of 150, 250 and 350 ms). Conditions were presented in a random sequence. The patients performed two blocks of 11 movements in each condition. The first three movements of each block were excluded for data analysis in order to report on steady-state tracking. As before [6,7], both eye and hand movements were distinguished into transport and target phases. The eye and hand transport phases corresponded to the primary saccadic and hand movement, respectively, starting when the velocity exceeded 50 mm/s and ending when velocity dropped below 50 mm/s. The target phases of the eye and hand immediately followed the transport phases and were defined as a fixed time window of 1 s to allow comparison between the conditions. For the hand transport phase, the duration and peak velocity were calculated as well as the initial error, defined as the mean distance between the target and hand position at the end of the transport phase. For both the eye and hand target phases, the additional path length was computed as the total length of the trajectory covered by the eye or hand in the target phase minus the straight distance between the position of eye and hand, respectively, at the start and the end of the target phase. The eye and hand additional path length have been shown to reflect the overall size of the fixational eye movements and overall tremor amplitude, respectively [5,7]. The study was conducted in accordance to the ethical standards set forth in the 1964 Declaration of Helsinki. Approval of the ethics committee and the informed consent of each patient were obtained before testing. Thirteen MS patients with intention tremor, with mean age of 49 (±12.2) years participated into the study. Patients with pyramidal signs and sensory loss in the upper limb or low visual acuity, restricted eye mobility and cognitive dysfunction interfering with the execution of the wrist step-tracking task, were not included. The mean disease duration was 13.9 years and the median score on the expanded disability status scale was six. All patients had bilateral intention tremor whose severity during the finger-to-nose test, rated using Fahn’s tremor rating scale, ranged between one and four. All subjects performed the task bilaterally, except four patients with unilateral extremely severe tremor preventing an adequate rhythmical execution of the task. As such, the patient group consisted of 22 arms. In addition, 14 age-matched healthy control subjects with mean age of 47.7 (±9.3) years were tested bilaterally to investigate the effects of delayed visual feedback on normal step-tracking performance. Data of control and patient groups were grouped according to the number of hands as in previous work [6–8,15]. This was done as the tremor severity of left and right hand are very likely caused by different brain lesions [9], and as left and right tremor severity in this patient sample was not significantly correlated. For each group, repeated measures analyses of variance were performed to test for differences between the conditions (1, 150, 250 and 350 ms). Bonferroni post hoc tests were computed to correct for multiple comparisons. The level of significance was set at P < 0.05. An overview of the actual and represented hand step-tracking performance of the control and patient groups is presented
P. Feys et al. / Neuroscience Letters 394 (2006) 17–21
19
Table 1 Overview of the actual and represented hand step-tracking performance of both patient and control groups in all conditions
Transport phase Duration (ms)
PTN CTRL
Peak velocity (mm/s)
PTN CTRL
Initial error (mm)
PTN CTRL
Target phase Additional path length (mm)
PTN CTRL
1 ms
150 ms
250 ms
350 ms
X (S.D.)
X (S.D.)
X (S.D.)
X (S.D.)
Actual hand Representation Actual hand Representation
597 (139) 597 (139) 845 (168) 845 (168)
621 (168) 723 (182) 839 (149) 930 (148)
646 (160) 851 (199) 839 (172) 978 (172)
659 (176) 899 (193) 839 (149) 882 (194)
Actual hand Representation Actual hand Representation
1864 (576) 1864 (576) 1354 (349) 1354 (349)
1813 (598) 1607 (492) 1360 (301) 1274 (275)
1723 (585) 1397 (424) 1358 (332) 1198 (265)
1747 (584) 1328 (391) 1369 (284) 1139 (203)
Actual hand Representation Actual hand Representation
−17.5 (9.9) −17.5 (9.9) −6 (2.3) −6 (2.3)
−15.3 (11.6) −10.4 (10.9) −6.6 (3.9) −3.9 (2.4)
−16.2 (9.1) −9.3 (8.8) −6.7 (3.9) −4.0 (2.9)
−13.9 (11) −5.9 (9.3) −8.7 (5.5) −4.2 (3.1)
Actual hand Representation Actual hand Representation
47.3 (70.8) 47.3 (70.8) 1.3 (1.4) 1.3 (1.4)
24.6 (25.8) 18.7 (21) 1.2 (1.2) 1.0 (0.8)
24.4 (36.2) 19.4 (42.7) 1.5 (2.2) 1.3 (1.2)
25.9 (37.8) 16.2 (26.2) 1.7 (2.1) 1.3 (0.9)
in Table 1. Results of the represented step-tracking, i.e., the (changed) visual feedback on the screen, were included to clarify the changes in actual step-tracking performance between different conditions. The results of the hand step-tracking performance of the patient group revealed a greater duration of the actual, and especially the represented transport phase when the averaging procedure was applied over a longer time window (F(3.63) = 4.69, P < 0.001 and F(3.63) = 63.8, P < 0.0001, respectively). However, it must be noted that the peak velocity of the actual hand performance was not significantly different between the conditions. The patient group initially undershot the target equally across conditions, while the represented initial hand error was smaller when the visual feedback of the hand position was changed (F(3.63) = 13,41, P < 0.0001). Most interestingly, both the actual (F(3.63) = 3.88, P < 0.05) and represented hand additional path length (F(3.63) = 6.68, P < 0.001) decreased significantly when the representation of the hand position was averaged over an extended time window compared with the 1 ms time window condition (see Fig. 2). Less tremor was visually represented in the conditions with changed visual feedback than actually
Fig. 2. Mean and standard error of the actual and represented hand additional path length of the patient group in all conditions.
recorded (F(1.21) = 24.68, P < 0.001). These findings suggest that the amplitude of tremor was significantly reduced by the delayed visual display of an averaged tremor amplitude. The eye additional eye path of the patient group was not significantly different between the conditions (1 ms, 56.2 mm; 150 ms, 51.1 mm; 250 ms, 49.4 mm; 350 ms, 21.7 mm). The hand step-tracking performance of the control group was not affected by the delayed feedback, except for a greater initial error in the 350 ms condition (F(3.81) = 3.6, p < 0.05). In line with the averaging procedure, the hand representation on the visual display was characterised by a slower duration of the transport phase (F(3.81) = 21.3, p < 0.0001) and slower peak velocity (F(3.81) = 12.1, p < 0.0001). The eye additional path length of the control subjects was not influenced by the delayed visual feedback (1 ms, 24.1 mm; 150 ms, 23.4 mm; 250 ms, 22.5 mm; 350 ms, 22.4 mm). The major finding of this study was the reduction in the tremor amplitude induced by manipulated visual feedback, indicating the visual dependence of intention tremor. This important finding could be explained by several factors. Firstly, the appliance of the averaging procedure over a larger time window led to a spatial decrease of the visually displayed initial tracking error and overall tremor amplitude. The actual overall tremor amplitude during visually guided tracking can be reduced as a consequence of reduced voluntary error correction movements in response to the decreased visually displayed tracking errors. A ‘gain’ mechanism may be involved: the smaller the perceived error, the smaller the voluntary correction movement and related involuntary tremor. The results of this study do not allow us to differentiate if the tremor reduction is mainly related to the visually decreased initial error, which is changed position information, or to the visually decreased tremor amplitude, which is changed motion or trajectory information. A recent study, however, showed that both position and motion feedback signals
20
P. Feys et al. / Neuroscience Letters 394 (2006) 17–21
are used for online control of discrete goal-directed movements [23]. Secondly, the averaging procedure also induced temporal delays in the visual display of the hand position during tracking, which may clarify the observed reduction of the overall tremor amplitude. The conduction time of the visuomotor feedback loop was close to 200 ms or more in normal subjects [19], and increased in MS patients with tremor [5,14,16]. As such, the delay in presenting visual feedback may have helped to react to errors more efficiently. A noteworthy observation is that there was no significant difference in tremor reduction over the different time windows of 150, 250 and 350 ms, although the delay increased when the size of the time window increased. It seems that once the delay passed a certain threshold, 150 ms in the present study, its effects on tremor amplitude were consistent. These consistent effects of different delays on tremor amplitude may seem to be in contrast with the studies in normal subjects, which reported that delays in providing visual feedback had a graded effect on visual adaptation [12,18]. Another difference is that the tracking errors in the present study were not increased in conditions with manipulated feedback, except for the 350 ms condition in the control group, whereas Foulkes and Miall [12] reported an increase proportional with the length of the delay. The different findings between studies may be explained by task and experimental design differences. The targets in our one-dimensional step-tracking task were stationary and thus with a predictable position, whereas unpredictable continuously two-dimensional moving targets were tracked in other studies [12,18]. Only a limited number of tracking trials were performed in each condition of the present study, perhaps not allowing more significant adaptation to the time delay to occur. It must also be noted that, in contrast to the studies using a fixed temporal delay in presenting visual feedback [12,24], the length of the delay in the present study was not precisely predictable as dependent on the movement characteristics, such as speed, direction and amplitude. In fact, MS patients with tremor did not adapt their primary hand movement to the delays as reflected by an unchanged peak velocity and actual initial error across conditions. The latter finding in the patient group also indicates that hand movement speed is unlikely to play a role in tremor reduction observed in the present study. Thirdly, one may argue that the spatial and temporal changes in the visual feedback of the hand position influenced the eye movements during tracking, in particular target fixation. Our previous study on eye–hand interaction in MS patients with tremor revealed that target fixation deficits during discrete step-tracking probably increased the severity of intention tremor [7], as they compromise the accurate location of the target relative to the hand position, hampering the tracking accuracy and thus evoking additional voluntary hand movements. With the delay and reduction of the visually displayed tremor, one could think that the target fixation would be more stable leading to a reduction of tremor amplitude. However, the results of the eye additional path length showed no differences between the conditions suggesting that fixational eye movements were unlikely be a factor contributing to the reduction in actual tremor amplitude.
We conclude that tremor amplitude was significantly reduced during the visually guided tracking tasks when the visual feedback of the hand position was changed by the averaging procedure. The results of the present study also provide support for specific rehabilitation interventions in MS tremor. Previous research showed that the handwriting and the mouse-driven PC-interaction in MS patients with intention tremor were significantly improved when visual feedback of the hand position was similarly manipulated [2,10]. The improved ability in conducting the computer mouse was likely to be related to a reduction of actual tremor amplitude. Acknowledgements Peter Feys is recipient of a postdoctoral fellowship from the Research Foundation (FWO)—Flanders. The technical assistance of Marc Beirinckx and Paul Meugens, and the contribution of Annick Janssens and Iris Dierickx is greatly appreciated. References [1] S.H. Alusi, J. Worthington, S. Glickman, P.G. Bain, A study of tremor in multiple sclerosis, Brain 124 (2001) 720–730. [2] C.A. Avizzano, F. Barbagli, P. Feys, M. Bergamasco, Adaptive filters for the suppression of tremor, in: IEEE International Conference: System, Man and Cybernetics, Nashville, TN, USA, 2000, pp. 1805–1810. [3] F. Debaere, N. Wenderoth, S. Sunaert, P. Van Hecke, S.P. Swinnen, Internal vs. external generation of movements: differential neural pathways involved in bimanual coordination performed in the presence or absence of augmented visual feedback, Neuroimage 19 (2003) 764– 776. [4] G. Deuschl, P. Bain, M. Brin, Consensus statement of the Movement Disorder Society on tremor. Ad hoc scientific committee, Mov. Disord. 13 (1998) 2–23. [5] P. Feys, W.F. Helsen, A. Lavrysen, B. Nuttin, P. Ketelaer, Intention tremor during manual aiming: a study of eye and hand movements, Mult. Scler. 9 (2003) 44–54. [6] P. Feys, W.F. Helsen, X. Liu, A. Lavrysen, V. Loontjens, B. Nuttin, P. Ketelaer, Effect of visual information on step-tracking movements in patients with intention tremor due to multiple sclerosis, Mult. Scler. 9 (2003) 492–502. [7] P. Feys, W. Helsen, X. Liu, A. Lavrysen, B. Nuttin, S.P. Swinnen, P. Ketelaer, Interactions between abnormal eye and hand movements in MS patients with intention tremor, Mov. Disord. 20 (2005) 705– 713. [8] P. Feys, W. Helsen, X. Liu, D. Mooren, H. Albrecht, B. Nuttin, P. Ketelaer, Effect of forearm cooling on intention tremor in multiple sclerosis, J. Neurol. Neurosurg. Psychiatry 76 (2005) 373–379. [9] P. Feys, F. Maes, B. Nuttin, W. Helsen, V. Malfait, G. Nagels, A. Lavrysen, X. Liu, Relationship between multiple sclerosis intention tremor severity and lesion load in the brainstem, Neuroreport 16 (2) (2005) 1379–1382. [10] P. Feys, A. Romberg, J. Ruutiainen, A. Davies-Smith, R. Jones, C.A. Avizzano, M. Bergamasco, P. Ketelaer, Assistive technology to improve PC interaction for people with intention tremor, J. Rehabil. Res. Dev. 38 (2001) 235–243. [11] P. Feys, A. Romberg, J. Ruutiainen, P. Ketelaer, Interference of upper limb tremor on daily life activities in people with multiple sclerosis, Occup. Ther. Health Care 17 (2004) 81–95. [12] A.J. Foulkes, R.C. Miall, Adaptation to visual feedback delays in a human manual tracking task, Exp. Brain Res. 131 (2000) 101–110. [13] M. Glickstein, How are visual areas of the brain connected to motor areas for the sensory guidance of the movement? Trends Cogn. Sci. 23 (2000) 613–617.
P. Feys et al. / Neuroscience Letters 394 (2006) 17–21 [14] X. Liu, H.A. Ingram, J.A. Palace, R.C. Miall, Dissociation of ‘online’ and ‘off-line’ visuomotor control of the arm by focal lesions in the cerebellum and brainstem, Neurosci. Lett. 264 (1999) 121–124. [15] X. Liu, C. Miall, T.Z. Aziz, J.A. Palace, P.N. Haggard, J.F. Stein, Analysis of action tremor and impaired control of movement velocity in multiple sclerosis during visually guided wrist-tracking tasks, Mov. Disord. 12 (1997) 992–999. [16] X. Liu, R.C. Miall, T.Z. Aziz, J.A. Palace, J.F. Stein, Distal versus proximal arm tremor in multiple sclerosis assessed by visually guided tracking tasks, J. Neurol. Neurosurg. Psychiatry 66 (1999) 43–47. [17] X. Liu, E. Robertson, R.C. Miall, Neuronal activity related to the visual representation of arm movements in the lateral cerebellar cortex, J. Neurophysiol. 89 (2003) 1223–1237. [18] R.C. Miall, D.J. Weir, J.F. Stein, Visuomotor tracking with delayed visual feedback, Neuroscience 16 (1985) 511–520. [19] R.C. Miall, D.J. Weir, J.F. Stein, Intermittency in human manual tracking tasks, J. Mot. Behav. 25 (1993) 53–63.
21
[20] H. Mitoma, Intention tremor exaggerated by visually guided movement, Eur. Neurol. 36 (1996) 177–178. [21] S.J. Pittock, R.L. McClelland, W.T. Mayr, M. Rodriguez, J.Y. Matsumoto, Prevalence of tremor in multiple sclerosis and associated disability in the Olmsted County population, Mov. Disord. 19 (2004) 1482–1485. [22] D.W. Reed, X. Liu, R.C. Miall, Online feedback control of human visually guided slow ramp tracking: effects of spatial separation of visual cues, Neurosci. Lett. 338 (2003) 209–212. [23] J.A. Saunders, D.C. Knill, Visual feedback control of hand movements, J. Neurosci. 24 (2004) 3223–3234. [24] J.L. Vercher, G.M. Gauthier, Oculo-manual coordination control: ocular and manual tracking of visual targets with delayed visual feedback of the hand motion, Exp. Brain Res. 90 (1992) 599–609. [25] D.M. Wolpert, R.C. Miall, M. Kawato, Internal models in the cerebellum, Trends Cogn. Sci. 2 (1998) 338–347.
The effect of changed visual feedback on intention tremor in multiple sclerosis Peter Feys a,c,∗ , Werner Helsen a , Martinus Buekers a , Tanja Ceux a , Elke Heremans a , Bart Nuttin b , Pierre Ketelaer c , Xuguang Liu d a
Katholieke Universiteit Leuven, Department of Biomedical Kinesiology, Motor Learning Laboratory, Tervuursevest 101, 3001 Leuven, Belgium b Katholieke Universiteit Leuven, Department of Neurosciences, Belgium c National Multiple Sclerosis Center, Melsbroek, Belgium d Imperial College, Department of Neurosciences, Charing Cross Hospital, UK Received 8 July 2005; received in revised form 7 September 2005; accepted 2 October 2005
Abstract In patients with multiple sclerosis (MS), intention tremor amplitude is enhanced during the visually guided compared to the memory guided motor tasks. In the present study, the effect of visual feedback on intention tremor was investigated during visually guided wrist step-tracking tasks. Specifically, visual feedback of the hand was provided either instantly or averaged over different time windows. Thirteen MS patients with intention tremor and 14 healthy controls performed the wrist step-tracking task, while the visual representation of the actual hand position was displayed instantly or averaged over time windows of 150, 250 and 350 ms. It has been found in the patient group that, in association with a decreased initial error and decreased tremor amplitude on the screen, the amplitude of the actual performed tremor also decreased when visual feedback was changed. The tremor reduction was not different between conditions with manipulated feedback, although delays in presenting visual feedback of the hand position increased when the time window was larger. The reduction in overall tremor amplitude was unlikely related to other factors, such as eye fixation deficits or the speed of the primary hand movement. These results suggest that hand tremor severity is dependent on the visual feedback of position and movement errors. © 2005 Elsevier Ireland Ltd. All rights reserved. Keywords: Intention tremor; Visual feedback; Multiple sclerosis; Step-tracking; Delay
Upper limb intention tremor, clinically defined as an increase in tremor amplitude towards the termination of a visually guided, goal-directed movement, is frequently encountered in multiple sclerosis (MS) [1,21]. This action tremor interferes with the performance of many daily life activities, such as drinking or mouse-driven PC interaction [10,11]. On neurophysiological level, intention tremor is related to the dysfunctioning of the cerebellum and/or afferent and efferent pathways [1,9]. The cerebellar system is heavily involved in the online control of visually guided, goal-directed movements, as the cerebellum is regarded as the neural locus of internal models that mutually connect sensory information and motor actions [3,13,17,25]. Previous research has shown that MS intention tremor is greatly enhanced during the visually guided in comparison to
∗
Corresponding author. Tel.: +32 16 329098. E-mail address: [email protected] (P. Feys).
0304-3940/$ – see front matter © 2005 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2005.10.010
the memory guided, goal-directed movements [6,15,20]. This robust finding was observed after complete occlusion of the target or movement cue during both continuous ramp-tracking with a moving target [15] and discrete step-tracking of stationary targets [6]. Also a partial occlusion of the movement cue, in the spatial area between or around the target positions, induced a reduction of tremor during step-tracking [6]. In healthy subjects, visual feedback control of slow ramp-tracking movements is dependent upon visually comparing both the target and limb positions to detect errors in the ongoing movement [22]. The tracking accuracy is then obtained by voluntary error correction movements. In MS patients, intention tremor is closely related to the execution of voluntary goal-directed movements [1]. The exaggeration of overall tremor amplitude during visually guided movements was caused likely by visually dependent voluntary error correction movements [6]. The present study extends the previous research on the influence of visual feedback on intention tremor in MS by changing the visual feedback
18
P. Feys et al. / Neuroscience Letters 394 (2006) 17–21
of the hand movement. The visual dependency of tremor amplitude on feedback of the ongoing movement is investigated by manipulating the visual representation of the hand movement during a wrist step-tracking task [6,7]. In addition to the hand movements, eye movements are also recorded as increased postsaccadic unsteady fixation was shown to influence the overall tremor amplitude in MS patients [7]. Eye and hand movements were simultaneously recorded using a combined eye-head tracker (66 Hz) and an angle encoder (200 Hz), respectively. The angle encoder was mounted on a wrist orthosis that allowed flexion and extension movements. The hand position was represented on a computer screen by an unfilled circle, which allowed us to apply dynamic changes to the visual feedback. Two stationary targets were displayed at the left and right side of the screen with an intertarget distance of 200 mm, corresponding to 40◦ wrist angle. The subjects were instructed to place the circle as accurately as possible at the target and to initiate the movement to the alternate target as soon as a beep was provided by a metronome (0.25 Hz or 2 s between two consecutive beeps). A wooden cover prevented the participant from direct sight of the actual wrist movements. Besides an instant visual display of the hand position, the visual feedback of the actual hand position was manipulated by representing the average of the tracking movement over preselected time windows. Specifically, the hand positions were averaged over windows of 150, 250, and 350 ms. The time windows were chosen based on the typical frequency range of intention tremor (3–6 Hz) [4] to cover one or more tremor cycles. As such, the averaging procedure visually “damped” the amplitude of the low-frequency tremor; however, also represented it with some time delay. For instance, if a patient’s tremor has a frequency of 4 Hz, one tremor cycle has a duration of 250 ms. In this case, tremor amplitude will be visually smaller when averaged over a time window of 250 ms, and be represented with a delay that is dependent on the movement characteristics, such as speed and amplitude (see Fig. 1 for illustration). Taken together, comparisons of the wrist step-tracking performance were made across
Fig. 1. Illustration of the actual hand movement (1 ms, black) of a patient case with marked tremor and its representation on the visual display when averaged over a time window of 150 ms (green), 250 ms (blue) and 350 ms (red).
four visual conditions: one without (time window of 1 ms) and three conditions with manipulation of the visual feedback (time windows of 150, 250 and 350 ms). Conditions were presented in a random sequence. The patients performed two blocks of 11 movements in each condition. The first three movements of each block were excluded for data analysis in order to report on steady-state tracking. As before [6,7], both eye and hand movements were distinguished into transport and target phases. The eye and hand transport phases corresponded to the primary saccadic and hand movement, respectively, starting when the velocity exceeded 50 mm/s and ending when velocity dropped below 50 mm/s. The target phases of the eye and hand immediately followed the transport phases and were defined as a fixed time window of 1 s to allow comparison between the conditions. For the hand transport phase, the duration and peak velocity were calculated as well as the initial error, defined as the mean distance between the target and hand position at the end of the transport phase. For both the eye and hand target phases, the additional path length was computed as the total length of the trajectory covered by the eye or hand in the target phase minus the straight distance between the position of eye and hand, respectively, at the start and the end of the target phase. The eye and hand additional path length have been shown to reflect the overall size of the fixational eye movements and overall tremor amplitude, respectively [5,7]. The study was conducted in accordance to the ethical standards set forth in the 1964 Declaration of Helsinki. Approval of the ethics committee and the informed consent of each patient were obtained before testing. Thirteen MS patients with intention tremor, with mean age of 49 (±12.2) years participated into the study. Patients with pyramidal signs and sensory loss in the upper limb or low visual acuity, restricted eye mobility and cognitive dysfunction interfering with the execution of the wrist step-tracking task, were not included. The mean disease duration was 13.9 years and the median score on the expanded disability status scale was six. All patients had bilateral intention tremor whose severity during the finger-to-nose test, rated using Fahn’s tremor rating scale, ranged between one and four. All subjects performed the task bilaterally, except four patients with unilateral extremely severe tremor preventing an adequate rhythmical execution of the task. As such, the patient group consisted of 22 arms. In addition, 14 age-matched healthy control subjects with mean age of 47.7 (±9.3) years were tested bilaterally to investigate the effects of delayed visual feedback on normal step-tracking performance. Data of control and patient groups were grouped according to the number of hands as in previous work [6–8,15]. This was done as the tremor severity of left and right hand are very likely caused by different brain lesions [9], and as left and right tremor severity in this patient sample was not significantly correlated. For each group, repeated measures analyses of variance were performed to test for differences between the conditions (1, 150, 250 and 350 ms). Bonferroni post hoc tests were computed to correct for multiple comparisons. The level of significance was set at P < 0.05. An overview of the actual and represented hand step-tracking performance of the control and patient groups is presented
P. Feys et al. / Neuroscience Letters 394 (2006) 17–21
19
Table 1 Overview of the actual and represented hand step-tracking performance of both patient and control groups in all conditions
Transport phase Duration (ms)
PTN CTRL
Peak velocity (mm/s)
PTN CTRL
Initial error (mm)
PTN CTRL
Target phase Additional path length (mm)
PTN CTRL
1 ms
150 ms
250 ms
350 ms
X (S.D.)
X (S.D.)
X (S.D.)
X (S.D.)
Actual hand Representation Actual hand Representation
597 (139) 597 (139) 845 (168) 845 (168)
621 (168) 723 (182) 839 (149) 930 (148)
646 (160) 851 (199) 839 (172) 978 (172)
659 (176) 899 (193) 839 (149) 882 (194)
Actual hand Representation Actual hand Representation
1864 (576) 1864 (576) 1354 (349) 1354 (349)
1813 (598) 1607 (492) 1360 (301) 1274 (275)
1723 (585) 1397 (424) 1358 (332) 1198 (265)
1747 (584) 1328 (391) 1369 (284) 1139 (203)
Actual hand Representation Actual hand Representation
−17.5 (9.9) −17.5 (9.9) −6 (2.3) −6 (2.3)
−15.3 (11.6) −10.4 (10.9) −6.6 (3.9) −3.9 (2.4)
−16.2 (9.1) −9.3 (8.8) −6.7 (3.9) −4.0 (2.9)
−13.9 (11) −5.9 (9.3) −8.7 (5.5) −4.2 (3.1)
Actual hand Representation Actual hand Representation
47.3 (70.8) 47.3 (70.8) 1.3 (1.4) 1.3 (1.4)
24.6 (25.8) 18.7 (21) 1.2 (1.2) 1.0 (0.8)
24.4 (36.2) 19.4 (42.7) 1.5 (2.2) 1.3 (1.2)
25.9 (37.8) 16.2 (26.2) 1.7 (2.1) 1.3 (0.9)
in Table 1. Results of the represented step-tracking, i.e., the (changed) visual feedback on the screen, were included to clarify the changes in actual step-tracking performance between different conditions. The results of the hand step-tracking performance of the patient group revealed a greater duration of the actual, and especially the represented transport phase when the averaging procedure was applied over a longer time window (F(3.63) = 4.69, P < 0.001 and F(3.63) = 63.8, P < 0.0001, respectively). However, it must be noted that the peak velocity of the actual hand performance was not significantly different between the conditions. The patient group initially undershot the target equally across conditions, while the represented initial hand error was smaller when the visual feedback of the hand position was changed (F(3.63) = 13,41, P < 0.0001). Most interestingly, both the actual (F(3.63) = 3.88, P < 0.05) and represented hand additional path length (F(3.63) = 6.68, P < 0.001) decreased significantly when the representation of the hand position was averaged over an extended time window compared with the 1 ms time window condition (see Fig. 2). Less tremor was visually represented in the conditions with changed visual feedback than actually
Fig. 2. Mean and standard error of the actual and represented hand additional path length of the patient group in all conditions.
recorded (F(1.21) = 24.68, P < 0.001). These findings suggest that the amplitude of tremor was significantly reduced by the delayed visual display of an averaged tremor amplitude. The eye additional eye path of the patient group was not significantly different between the conditions (1 ms, 56.2 mm; 150 ms, 51.1 mm; 250 ms, 49.4 mm; 350 ms, 21.7 mm). The hand step-tracking performance of the control group was not affected by the delayed feedback, except for a greater initial error in the 350 ms condition (F(3.81) = 3.6, p < 0.05). In line with the averaging procedure, the hand representation on the visual display was characterised by a slower duration of the transport phase (F(3.81) = 21.3, p < 0.0001) and slower peak velocity (F(3.81) = 12.1, p < 0.0001). The eye additional path length of the control subjects was not influenced by the delayed visual feedback (1 ms, 24.1 mm; 150 ms, 23.4 mm; 250 ms, 22.5 mm; 350 ms, 22.4 mm). The major finding of this study was the reduction in the tremor amplitude induced by manipulated visual feedback, indicating the visual dependence of intention tremor. This important finding could be explained by several factors. Firstly, the appliance of the averaging procedure over a larger time window led to a spatial decrease of the visually displayed initial tracking error and overall tremor amplitude. The actual overall tremor amplitude during visually guided tracking can be reduced as a consequence of reduced voluntary error correction movements in response to the decreased visually displayed tracking errors. A ‘gain’ mechanism may be involved: the smaller the perceived error, the smaller the voluntary correction movement and related involuntary tremor. The results of this study do not allow us to differentiate if the tremor reduction is mainly related to the visually decreased initial error, which is changed position information, or to the visually decreased tremor amplitude, which is changed motion or trajectory information. A recent study, however, showed that both position and motion feedback signals
20
P. Feys et al. / Neuroscience Letters 394 (2006) 17–21
are used for online control of discrete goal-directed movements [23]. Secondly, the averaging procedure also induced temporal delays in the visual display of the hand position during tracking, which may clarify the observed reduction of the overall tremor amplitude. The conduction time of the visuomotor feedback loop was close to 200 ms or more in normal subjects [19], and increased in MS patients with tremor [5,14,16]. As such, the delay in presenting visual feedback may have helped to react to errors more efficiently. A noteworthy observation is that there was no significant difference in tremor reduction over the different time windows of 150, 250 and 350 ms, although the delay increased when the size of the time window increased. It seems that once the delay passed a certain threshold, 150 ms in the present study, its effects on tremor amplitude were consistent. These consistent effects of different delays on tremor amplitude may seem to be in contrast with the studies in normal subjects, which reported that delays in providing visual feedback had a graded effect on visual adaptation [12,18]. Another difference is that the tracking errors in the present study were not increased in conditions with manipulated feedback, except for the 350 ms condition in the control group, whereas Foulkes and Miall [12] reported an increase proportional with the length of the delay. The different findings between studies may be explained by task and experimental design differences. The targets in our one-dimensional step-tracking task were stationary and thus with a predictable position, whereas unpredictable continuously two-dimensional moving targets were tracked in other studies [12,18]. Only a limited number of tracking trials were performed in each condition of the present study, perhaps not allowing more significant adaptation to the time delay to occur. It must also be noted that, in contrast to the studies using a fixed temporal delay in presenting visual feedback [12,24], the length of the delay in the present study was not precisely predictable as dependent on the movement characteristics, such as speed, direction and amplitude. In fact, MS patients with tremor did not adapt their primary hand movement to the delays as reflected by an unchanged peak velocity and actual initial error across conditions. The latter finding in the patient group also indicates that hand movement speed is unlikely to play a role in tremor reduction observed in the present study. Thirdly, one may argue that the spatial and temporal changes in the visual feedback of the hand position influenced the eye movements during tracking, in particular target fixation. Our previous study on eye–hand interaction in MS patients with tremor revealed that target fixation deficits during discrete step-tracking probably increased the severity of intention tremor [7], as they compromise the accurate location of the target relative to the hand position, hampering the tracking accuracy and thus evoking additional voluntary hand movements. With the delay and reduction of the visually displayed tremor, one could think that the target fixation would be more stable leading to a reduction of tremor amplitude. However, the results of the eye additional path length showed no differences between the conditions suggesting that fixational eye movements were unlikely be a factor contributing to the reduction in actual tremor amplitude.
We conclude that tremor amplitude was significantly reduced during the visually guided tracking tasks when the visual feedback of the hand position was changed by the averaging procedure. The results of the present study also provide support for specific rehabilitation interventions in MS tremor. Previous research showed that the handwriting and the mouse-driven PC-interaction in MS patients with intention tremor were significantly improved when visual feedback of the hand position was similarly manipulated [2,10]. The improved ability in conducting the computer mouse was likely to be related to a reduction of actual tremor amplitude. Acknowledgements Peter Feys is recipient of a postdoctoral fellowship from the Research Foundation (FWO)—Flanders. The technical assistance of Marc Beirinckx and Paul Meugens, and the contribution of Annick Janssens and Iris Dierickx is greatly appreciated. References [1] S.H. Alusi, J. Worthington, S. Glickman, P.G. Bain, A study of tremor in multiple sclerosis, Brain 124 (2001) 720–730. [2] C.A. Avizzano, F. Barbagli, P. Feys, M. Bergamasco, Adaptive filters for the suppression of tremor, in: IEEE International Conference: System, Man and Cybernetics, Nashville, TN, USA, 2000, pp. 1805–1810. [3] F. Debaere, N. Wenderoth, S. Sunaert, P. Van Hecke, S.P. Swinnen, Internal vs. external generation of movements: differential neural pathways involved in bimanual coordination performed in the presence or absence of augmented visual feedback, Neuroimage 19 (2003) 764– 776. [4] G. Deuschl, P. Bain, M. Brin, Consensus statement of the Movement Disorder Society on tremor. Ad hoc scientific committee, Mov. Disord. 13 (1998) 2–23. [5] P. Feys, W.F. Helsen, A. Lavrysen, B. Nuttin, P. Ketelaer, Intention tremor during manual aiming: a study of eye and hand movements, Mult. Scler. 9 (2003) 44–54. [6] P. Feys, W.F. Helsen, X. Liu, A. Lavrysen, V. Loontjens, B. Nuttin, P. Ketelaer, Effect of visual information on step-tracking movements in patients with intention tremor due to multiple sclerosis, Mult. Scler. 9 (2003) 492–502. [7] P. Feys, W. Helsen, X. Liu, A. Lavrysen, B. Nuttin, S.P. Swinnen, P. Ketelaer, Interactions between abnormal eye and hand movements in MS patients with intention tremor, Mov. Disord. 20 (2005) 705– 713. [8] P. Feys, W. Helsen, X. Liu, D. Mooren, H. Albrecht, B. Nuttin, P. Ketelaer, Effect of forearm cooling on intention tremor in multiple sclerosis, J. Neurol. Neurosurg. Psychiatry 76 (2005) 373–379. [9] P. Feys, F. Maes, B. Nuttin, W. Helsen, V. Malfait, G. Nagels, A. Lavrysen, X. Liu, Relationship between multiple sclerosis intention tremor severity and lesion load in the brainstem, Neuroreport 16 (2) (2005) 1379–1382. [10] P. Feys, A. Romberg, J. Ruutiainen, A. Davies-Smith, R. Jones, C.A. Avizzano, M. Bergamasco, P. Ketelaer, Assistive technology to improve PC interaction for people with intention tremor, J. Rehabil. Res. Dev. 38 (2001) 235–243. [11] P. Feys, A. Romberg, J. Ruutiainen, P. Ketelaer, Interference of upper limb tremor on daily life activities in people with multiple sclerosis, Occup. Ther. Health Care 17 (2004) 81–95. [12] A.J. Foulkes, R.C. Miall, Adaptation to visual feedback delays in a human manual tracking task, Exp. Brain Res. 131 (2000) 101–110. [13] M. Glickstein, How are visual areas of the brain connected to motor areas for the sensory guidance of the movement? Trends Cogn. Sci. 23 (2000) 613–617.
P. Feys et al. / Neuroscience Letters 394 (2006) 17–21 [14] X. Liu, H.A. Ingram, J.A. Palace, R.C. Miall, Dissociation of ‘online’ and ‘off-line’ visuomotor control of the arm by focal lesions in the cerebellum and brainstem, Neurosci. Lett. 264 (1999) 121–124. [15] X. Liu, C. Miall, T.Z. Aziz, J.A. Palace, P.N. Haggard, J.F. Stein, Analysis of action tremor and impaired control of movement velocity in multiple sclerosis during visually guided wrist-tracking tasks, Mov. Disord. 12 (1997) 992–999. [16] X. Liu, R.C. Miall, T.Z. Aziz, J.A. Palace, J.F. Stein, Distal versus proximal arm tremor in multiple sclerosis assessed by visually guided tracking tasks, J. Neurol. Neurosurg. Psychiatry 66 (1999) 43–47. [17] X. Liu, E. Robertson, R.C. Miall, Neuronal activity related to the visual representation of arm movements in the lateral cerebellar cortex, J. Neurophysiol. 89 (2003) 1223–1237. [18] R.C. Miall, D.J. Weir, J.F. Stein, Visuomotor tracking with delayed visual feedback, Neuroscience 16 (1985) 511–520. [19] R.C. Miall, D.J. Weir, J.F. Stein, Intermittency in human manual tracking tasks, J. Mot. Behav. 25 (1993) 53–63.
21
[20] H. Mitoma, Intention tremor exaggerated by visually guided movement, Eur. Neurol. 36 (1996) 177–178. [21] S.J. Pittock, R.L. McClelland, W.T. Mayr, M. Rodriguez, J.Y. Matsumoto, Prevalence of tremor in multiple sclerosis and associated disability in the Olmsted County population, Mov. Disord. 19 (2004) 1482–1485. [22] D.W. Reed, X. Liu, R.C. Miall, Online feedback control of human visually guided slow ramp tracking: effects of spatial separation of visual cues, Neurosci. Lett. 338 (2003) 209–212. [23] J.A. Saunders, D.C. Knill, Visual feedback control of hand movements, J. Neurosci. 24 (2004) 3223–3234. [24] J.L. Vercher, G.M. Gauthier, Oculo-manual coordination control: ocular and manual tracking of visual targets with delayed visual feedback of the hand motion, Exp. Brain Res. 90 (1992) 599–609. [25] D.M. Wolpert, R.C. Miall, M. Kawato, Internal models in the cerebellum, Trends Cogn. Sci. 2 (1998) 338–347.