Presented at NACOB 98: North American Congress on Biomechanics Canadian Society for Biomechanics - American Society of Biomechanics University of Waterloo Waterloo, Ontario, Canada August 14-18, 1998

---

COORDINATION BETWEEN POSTURE AND GOAL-DIRECTED ARM MOVEMENTS
WHEN THE TARGET POSITION CHANGES

O. Martin ¹ , N. Teasdale ² , M. Simoneau ² , C. Bourdin ³ , M. Fleury ² , C. Bard ^2
1 Laboratoire GAM, Université de Bourgogne, 21000 Dijon, France
² Division de Kinésiologie, Université Laval, Québec, Canada G1K 7P4
³ UMR Mouvement et Perception, Université de la Méditerranée, 13000 Marseille, France

INTRODUCTION

Most voluntary upper-limb movements are performed with postural anticipatory responses (Bouisset & Zattara, 1981; Massion, 1992). Little is known, however, as to how these postural responses are regulated when goal-directed pointing or reaching movements are performed from an upright standing position. The aim of this experiment was to examine the nature of the on-line dynamic functional link between posture and accurate hand pointings when rapid visuomanual corrections are required.

REVIEW AND THEORY

The real-time control processes of arm movement directed to displaced visual targets have been studied extensively (e.g. Pélisson et al., 1986). In most of these experiments, subjects performed the task while seated and there was no specific postural constraint. On the other hand, upper limb movements without specific spatial goal are prececed by anticipatory postural commands that presumably compensate for the disequilibrium created by the focal movement. Therefore, goal-directed movements may impose additional programming and control constraints upon the postural command. To examine the relationship between postural control and goal-directed arm movements, subjects performed rapid hand pointings from a standing position to visual targets that could change position at hand movement onset.

PROCEDURES

Seven volunteers aged 24 to 36 (all males and right-handed) participated on a voluntary basis. They all gave informed consent. In the initial position, the right hand was in contact with the lower extremity of the sternum. The task required pointing with the right hand to one of two targets (light emitting diodes) positioned along the sagittal axis. The two targets were 20 cm apart, 10 cm ahead and 10 cm beyond a distance corresponding to the length of the right arm extended in a pointing position. In a control condition (n=20 trials), after a ready signal, one of the two targets was light up and the subject had to point as fast and as accurately as possible to the target. In a step-target condition (n=80 trials), a visual perturbation occurred randomly in 33% of the trials at hand movement onset, turning off the initial target and simultaneously lighting up the other one. In the present paper, trials directed initially to the farthest target are presented. The 3D kinematics of arm and postural movements were obtained with a Selspot II system. Markers were placed on the right ankle, knee, ear lobe, index, and both hips and shoulders. Flexor and extensor EMG activity of arm (right biceps and triceps) and trunk (left tensor fasciae latae and right semitendinosus) muscles were recorded with surface electrodes.

RESULTS

The analyses focused on the comparison between perturbed trials (step-target) and non-perturbed trials. In the latter case, the specific effects of step-target uncertainty were also examined by comparing trials in the control condition vs. non-perturbed trials in the step-target condition.

Visuomanual pointing analysis. Overall, less than 5% of the trials were rejected because of a lack of pointing accuracy at target contact (radial error greater than 1.5 cm). Movement times were longer for the step-target trials (increase of 180 ms) and the effect was mainly a consequence of an additional backward sub-movement following an initial peak deceleration (top panel of Figure 1). This is observed in the speed-time profils (middle panel).

Figure 1: Hand lateral trajectory, speed, and hip angle. Mean data for one subject.

Postural analysis. The initial standing position before movement onset (as measured from the hip and ankle angle) was not different across conditions (on average, hip:169 deg; ankle: 90 deg). During arm movement, the hip angle was more important (bottom panel of Figure 1) and varied more in the unperturbed and step-target conditions. Interestingly, step-target trials were characterized by a hip angle adaptation before changes could be observed in the hand kinematics (see arrow in bottom panel of Figure 1 vs. middle panel). Coordination between focal and postural motor commands. Figure 2 shows an increased biceps (focal) and semitendinosus (postural) activity for the unperturbed and step-target conditions. A tight synchronization between both muscles is also observed in these conditions.

Figure 2: Biceps and semitendinosus EMG activity. Identical arbitrary unit for each panel. Mean data for the same subject as Figure 1.

DISCUSSION

Central commands controlling posture and arm movement are prepared to quickly react to a visual perturbation. When a step-target was presented, both commands were synchronized at the muscular level but the global modifications were observed first at the postural level. This suggests an on-line synergy between the two commands and that accurate goal-directed perturbed movements are linked to modifications at the postural level.

REFERENCES

Bouisset S. & Zattara M. Neurosci Lett 22: 263-270, 1981.

Massion J. Prog in Neurobiol 38: 35-56, 1992.

Pélisson D, Prablanc C, Goodale MA, Jeannerod M. Exp Br Res 62: 303-311, 1986.

ACKNOWLEDGMENTS

Study supported in part by the " programmes scientifiques du Conseil Régional de Bourgogne, Dijon, France ", NSERC-Canada, and FCAR-Québec.