Tasks that are easy when performed in isolation become difficult when performed simultaneously in the upper and/or lower limbs. This observation points to basic CNS constraints in the organization of patterns of interlimb coordination. The present studies provide evidence for the existence of two basic coordinative constraints whose effects may be additive under certain conditions. On one hand, the egocentric constraint denotes a general preference for moving the limbs toward or away from the longitudinal axis of the body in a symmetrical fashion and is of primary importance during the coordination of homologous limbs. On the other hand, the allocentric constraint refers to a general preference to move the limbs in the same direction in extrinsic space and pertains to the coordination of nonhomologous limbs (eg., various combinations of the upper and lower limbs). In the present context, constraints are considered as expressions of basic features of CNS operation that give way to preferred coordination patterns to which the system is naturally drawn or biased. The identification and description of these constraints is considered of critical importance to obtain a better understanding of the control of coordination patterns.

Strategies used by the CNS to optimize arm movements in terms of speed, accuracy, and resistance to fatigue remain largely unknown. A hypothesis is studied that the CNS exploits biomechanical properties of multijoint limbs to increase efficiency of movement control. To test this notion, a novel free-stroke drawing task was used that instructs subjects to make straight strokes in as many different directions as possible in the horizontal plane through rotations of the elbow and shoulder joints. Despite explicit instructions to distribute strokes uniformly, subjects showed biases to move in specific directions. These biases were associated with a tendency to perform movements that included active motion at one joint and largely passive motion at the other joint, revealing a tendency to minimize intervention of muscle torque for regulation of the effect of interaction torque. Other biomechanical factors, such as inertial resistance and kinematic manipulability, were unable to adequately account for these significant biases. Also, minimizations of jerk, muscle torque change, and sum of squared muscle torque were analyzed; however, these cost functions failed to explain the observed directional biases. Collectively, these results suggest that knowledge of biomechanical cost functions regarding interaction torque (IT) regulation is available to the control system. This knowledge may be used to evaluate potential movements and to select movement of "low cost." The preference to reduce active regulation of interaction torque suggests that, in addition to muscle energy, the criterion for movement cost may include neural activity required for movement control. I N T R O D U C T I O NDemands of daily living promote optimization of movement characteristics, such as speed and accuracy, while minimizing effort for movement production. How this optimization is achieved has been a focus of extensive research in the area of optimal control of human movements. Various cost functions have been proposed (Todorov 2004); however, it is difficult to ascertain what is actually being optimized, as well as how this optimization process is organized. We hypothesize that the CNS exploits biomechanical properties of the limbs to increase efficiency of movement control. The study specifically focuses on biomechanical factors that influence performance of multijoint arm movements. Three such factors have been recognized: interaction torque (IT), inertial resistance, and kinematic manipulability. IT results from mechanical influence of arm segments on each other during motion (Hollerbach and Flash 1982). Inertial resistance characterizes muscle effort necessary to produce a given acceleration of the arm endpoint (Hogan 1985). Kinematic manipulability characterizes angular velocity at the joints required to produce a given endpoint velocity (Yoshikawa 1985(Yoshikawa , 1990.To produce goal-directed movements, muscular control must be adjusted to all these factors. Each factor depends on movement direction, thus imposing differential demands for...

This article presents a theoretical generalization of recent experimental findings accumulated in support of two concepts of inter-segmental dynamics regulation during multi-joint movements. The concepts are the internal model of inter-segmental dynamics and the leading joint hypothesis (LJH). The internal model of limb dynamics is a well-established interpretation of feed-forward control. Recent experiments have generated new information about the organization of the internal model and its role in regulation of inter-segmental dynamics. The LJH, which proposes a simplified principle of the regulation of inter-segmental dynamics, is at the beginning stage of development. This paper outlines major results obtained in these two research directions and demonstrates that the two groups of findings complement and augment each other, suggesting a simple and robust hierarchical strategy of multi-joint movement control that exploits specific mechanical properties of human limbs.

No abstract

Characteristics of control at the shoulder and elbow during nine types of drawing movements were studied in the present work. The task was to repetitively track a template, depicted on a horizontal table, with the index finger at a cyclic frequency of 1.5 Hz. The templates were a circle, four ovals and four lines of different orientations. The wrist was immobilized and the movement consisted of rotations at the shoulder and elbow joints. The studied movements varied in a wide range with respect to the amplitude of elbow and shoulder movements and relative phase between them. Kinetic analysis included analysis of torque signs, impulses, and timing. It demonstrated that the role of muscle torque in movement production was different at the two joints. During eight out of the nine movement types, the muscle torque at the shoulder accelerated and decelerated this joint and almost completely coped with the influence of the interactive torque arising from elbow motion. Conversely, interactive torque generated by shoulder motion played a dominant role in elbow acceleration and deceleration, whereas muscle torque at the elbow adjusted passive elbow movement to the various template shapes. EMG data were in agreement with the conclusions made from the kinetic analysis. Collectively, these data support the hypothesis that the two joints have different functions in movement production. The shoulder creates a foundation for motion of the entire arm through the interactive torque, and the elbow serves as a fine-tuner of the end-point movement. Control of the shoulder was similar across the eight movement types and the differences in the end-point path were provided by variations in elbow control. The two joints exchanged roles during one movement type, namely, drawing the line tilted right. During this movement, the elbow musculature generated motion at this joint and the shoulder musculature counteracted mechanical influence of this motion on the shoulder position. The findings suggest that during drawing movements, the control strategy exploits intersegmental dynamics of the shoulder-elbow mechanical linkage.

Irregularities in the velocity profile near the end of pointing movements have been interpreted as corrective submovements whose purpose is to provide accuracy of pointing to the target. The purpose of the present study was to investigate whether two additional factors related to biomechanical properties of the arm also cause submovements. First, motion termination and stabilization of the limb in the final position required by a discrete pointing task may contribute to submovements. Second, inaccurate regulation of interactive torque at the joints may also cause submovements. To investigate the contributions of these two biomechanical factors and the traditionally considered factor of pointing accuracy, the incidence of submovements was analyzed during three types of experimental manipulations. In addition to target size manipulations (small and large targets), conditions for motion termination were manipulated by examining discrete movements (which terminated at the target) and reciprocal movements (which reversed direction without dwelling on the target). Interaction torques were varied by using targets that require different shoulder-elbow coordination patterns. Submovements were detected in 41% of all analyzed movements. Data supported influences from the accuracy and motion termination factors but not from the interactive torque regulation factor on submovement incidence. Gross submovements were associated with motion termination; fine submovements primarily with accuracy demands. These findings and the analysis of temporal movement characteristics suggest that motion termination is an extra movement component that makes control of discrete movements different to control of reciprocal movements. Implications of the findings to a noise-related interpretation of Fitts' law are discussed. The study emphasizes the influence of arm biomechanics on endpoint kinematics.

The generalizability of the preferred in-phase and anti-phase coordination modes under isofrequency conditions to bimanual patterns with a 2:1 frequency ratio was studied. Experiment 1 dealt with spontaneously emerging coordination modes and showed that all participants converged to a similar relative phasing pattern, characterized by an alternation between synchronization of the same and opposite relative peak limb positions. This suggests that movement reversals were exploited as intermittent loci of control during multifrequency tasks. Experiment 2 involved the acquisition of a 2:1 ratio with a 90° phase offset and demonstrated the powerful effect of real-time visual relative motion feedback on performance. Removal of this augmented feedback source resulted in a deterioration of the coordination pattern, accompanied by a regression to the aforementioned spontaneous coordination modes.

The present paper focused on the role of mechanical factors arising from the multijoint structure of the musculoskeletal system and their use in the control of different patterns of cyclical elbow-wrist movements. Across five levels of cycling frequency (from 0.45 Hz up to 3.05 Hz), three movement patterns were analyzed: (1) unidirectional, including rotations at the elbow and wrist in the same direction; (2) bidirectional, with rotation at the joints in opposite directions, and (3) free-wrist pattern, which is characterized by alternating flexions and extensions at the elbow with the wrist relaxed. Angular position of both joints and electromyographic activity of biceps, triceps, the wrist flexor, and the wrist extensor were recorded. It was demonstrated that control at the elbow was principally different from control at the wrist. Elbow control in all three patterns was similar to that typically observed during single-joint movements: elbow accelerations-decelerations resulted from alternating activity of the elbow flexor and extensor and were largely independent of wrist motion at all frequency plateaus. The elbow muscles were responsible not only for the elbow movement, but also for the generation of interactive torques that played an important role in wrist control. There were two types of interactive torques exerted at the wrist: inertial torque arising from elbow motion and restraining torque arising from physical limits imposed on wrist rotation. These interactive torques were the primary source of wrist motion, whereas the main function of wrist-muscle activity was to intervene with the interactive effects and to adjust the wrist movement to comply with the required coordination pattern. The unidirectional pattern was more in agreement with interactive effects than the bidirectional pattern, thus causing their differential difficulty at moderate cycle frequencies. When cycling frequency was further increased, both the unidirectional and bidirectional movements lost their individual features and acquired features of the free-wrist pattern. The deterioration of the controlled patterns at high cycling frequencies suggests a crucial role for proprioceptive information in wrist control. These results are supportive of a hierarchical organization of control with respect to elbow-wrist coordination, during which the functions of control at the elbow and wrist are principally different: the elbow muscles generate movement of the whole linkage and the wrist muscles produce corrections of the movement necessary to fulfill the task.