Increased velocity feedback gains in the presence of sensory noise can explain paradoxical changes in trunk motor control related to back pain.

Literature reports paradoxical findings regarding effects of low-back pain (LBP) on trunk motor control. Compared to healthy individuals, patients with LBP, especially those with high pain-related anxiety, showed stronger trunk extensor reflexes and more resistance against perturbations. On the other hand, LBP patients and especially those with high pain-related anxiety showed decreased precision in unperturbed trunk movement and posture. These paradoxical effects might be explained by arousal potentially increasing average and variance of muscle spindle firing rates. Increased average firing rates could increase resistance against perturbations, but increased variance could decrease precision. We performed a simulation study to test this hypothesis. We modeled the trunk as a 2D inverted pendulum, stabilized by two antagonistic Hill-type muscles, based on their open-loop muscle activation dependent intrinsic stiffness and damping and through 25 ms-delayed, noisy contractile element length and velocity feedback. Reference feedback gains and sensory noise levels were tuned based on previously reported experimental data. We assessed the effect of increasing feedback gains on precision of trunk orientation at different perturbation magnitudes and assessed sensitivity of the effects to open-loop muscle stimulation and noise levels. At low perturbation magnitudes, increasing reflex gains consistently caused an increase in the variance of trunk orientation. At larger perturbation magnitudes, increasing reflex gains consistently caused a decrease in the variance of trunk orientation. Our results support the notion that LBP and related anxiety may increase reflex gains, resulting in an increase in the average and variance of spindle afference, which in turn increase resistance against perturbations and decrease movement precision.

Limiting radial pedal forces greatly reduces maximal power output and efficiency in sprint cycling: an optimal control study.

A cyclist's performance depends critically on the generated average mechanical power output (AMPO). The instantaneous mechanical power output equals the product of crank angular velocity, crank length, and the tangential pedal force. Radial pedal forces do not contribute to mechanical power. It has been suggested that radial pedal forces arise from suboptimal pedaling technique and that limiting these would increase AMPO and efficiency. Here, we presented an optimal control musculoskeletal model of a cyclist (consisting of five segments driven by nine Hill-type muscle-tendon units) to predict maximal AMPO during sprint cycling at different levels of allowed radial pedal forces. Our findings showed that limiting radial pedal forces has a detrimental effect on maximal AMPO; it dropped from 1,115 W without a limit on radial forces to 528 W when no radial forces were allowed (both at 110 rpm). We explained that avoiding radial pedal forces causes ineffective use of muscles: muscles deliver less positive power and have a higher muscle power dissipation ratio (average mechanical power dissipated per unit of average positive power delivered). We concluded that radial pedal forces are an unavoidable by-product when optimizing for maximal AMPO and that limiting these leads to a performance decrease.NEW & NOTEWORTHY In the literature, but also in the "cycling field" [e.g., trainers, coaches, and (professional) cyclists], it is often suggested that trying to limit/avoid radial pedal forces enhances cycling technique and with that maximal average power output and efficiency. In this paper, we introduce an optimal control model of a human cyclists (consisting of five segments and driven by nine Hill-type muscle-tendon complex models). With that we not only show, but also explain why limiting radial forces is a bad idea: it will decrease maximal attainable AMPO and will decrease efficiency.

The Metabolic Cost of Walking in healthy young and older adults - A Systematic Review and Meta Analysis.

The Metabolic Cost of Walking (MCoW) is an important variable of daily life that has been studied extensively. Several studies suggest that MCoW is higher in Older Adults (OA) than in Young Adults (YA). However, it is difficult to compare values across studies due to differences in the way MCoW was expressed, the units in which it was reported and the walking speed at which it was measured. To provide an overview of MCoW in OA and YA and to investigate the quantitative effect of age on MCoW, we have conducted a literature review and performed two meta-analyses. We extracted data on MCoW in healthy YA (18-41 years old) and healthy OA (≥59 years old) and calculated, if not already reported, the Gross (GCoW) and Net MCoW (NCoW) in J/kg/m. If studies reported MCoW measured at multiple speeds, we selected those values for YA and OA at which MCoW was minimal. All studies directly comparing YA and OA were selected for meta-analyses. From all studies reviewed, the average GCoW in YA was 3.4 ± 0.4 J/kg/m and 3.8 ± 0.4 J/kg/m in OA (~12% more in OA), and the average NCoW in YA was 2.4 ± 0.4 J/kg/m and 2.8 ± 0.5 J/kg/m in OA (~17% more in OA). Our meta-analyses indicated a statistically significant elevation of both GCoW and NCoW (p < 0.001) for OA. In terms of GCoW, OA expended about 0.3 J/kg/m more metabolic energy than YA and about 0.4 J/kg/m more metabolic energy than YA in terms of NCoW. Our study showed a statistically significant elevation in MCoW of OA over YA. However, from the literature it is unclear if this elevation is directly caused by age or due to an interaction between age and methodology. We recommend further research comparing MCoW in healthy OA and YA during "natural" over-ground walking and treadmill walking, after sufficient familiarization time.

Metabolic Cost of Activation and Mechanical Efficiency of Mouse Soleus Muscle Fiber Bundles During Repetitive Concentric and Eccentric Contractions.

Currently available data on the energetics of isolated muscle preparations are based on bouts of less than 10 muscle contractions, whereas metabolic energy consumption is mostly relevant during steady state tasks such as locomotion. In this study we quantified the energetics of small fiber bundles of mouse soleus muscle during prolonged (2 min) series of contractions. Bundles (N = 9) were subjected to sinusoidal length changes, while measuring force and oxygen consumption. Stimulation (five pulses at 100 Hz) occurred either during shortening or during lengthening. Movement frequency (2-3 Hz) and amplitude (0.25-0.50 mm; corresponding to ± 4-8% muscle fiber strain) were close to that reported for mouse soleus muscle during locomotion. The experiments were performed at 32°C. The contributions of cross-bridge cycling and muscle activation to total metabolic energy expenditure were separated using blebbistatin. The mechanical work per contraction cycle decreased sharply during the first 10 cycles, emphasizing the importance of prolonged series of contractions. The mean ± SD fraction of metabolic energy required for activation was 0.37 ± 0.07 and 0.56 ± 0.17 for concentric and eccentric contractions, respectively (both 0.25 mm, 2 Hz). The mechanical efficiency during concentric contractions increased with contraction velocity from 0.12 ± 0.03 (0.25 mm 2 Hz) to 0.15 ± 0.03 (0.25 mm, 3 Hz) and 0.16 ± 0.02 (0.50 mm, 2 Hz) and was -0.22 ± 0.08 during eccentric contractions (0.25 mm, 2 Hz). The percentage of type I fibers correlated positively with mechanical efficiency during concentric contractions, but did not correlate with the fraction of metabolic energy required for activation.

Inverse dynamics of mechanical multibody systems: An improved algorithm that ensures consistency between kinematics and external forces.

Inverse dynamics is a technique in which measured kinematics and, possibly, external forces are used to calculate net joint torques in a rigid body linked segment model. However, kinematics and forces are usually not consistent due to incorrect modelling assumptions and measurement errors. This is commonly resolved by introducing 'residual forces and torques' which compensate for this problem, but do not exist in reality. In this study a constrained optimization algorithm is proposed that finds the kinematics that are mechanically consistent with measured external forces and mimic the measured kinematics as closely as possible. The algorithm was tested on datasets containing planar kinematics and ground reaction forces obtained during human walking at three velocities (0.8 m/s, 1.25 and 1.8 m/s). Before optimization, the residual force and torque were calculated for a typical example. Both showed substantial values, indicating the necessity of developing a mechanically consistent algorithm. The proposed optimization algorithm converged to a solution in which the residual forces and torques were zero, without changing the ground reaction forces and with only minor changes to the measured kinematics. When using a rigid body approach, our algorithm ensures a consistent description of forces and kinematics, thereby improving the validity of calculated net joint torque and power values.

Searching for strategies to reduce the mechanical demands of the sit-to-stand task with a muscle-actuated optimal control model.

BACKGROUND: The sit-to-stand task, which involves rising unassisted from sitting on a chair to standing, is important in daily life. Many people with muscle weakness, reduced range of motion or loading-related pain in a particular joint have difficulty performing the task. How should a person suffering from such impairment best perform the sit-to-stand task and, in the case of pain in a particular joint, with reduced loading of that joint? METHODS: We developed a musculoskeletal model with reference parameter values based on properties of healthy strong subjects. The model's muscle stimulation-time input was optimized using direct collocation to find strategies that yielded successful sit-to-stand task performance with minimum 'control effort' for the reference set and modified sets of parameter values, and with constraints on tibiofemoral compression force. FINDINGS: The sit-to-stand task could be performed successfully and realistically by the reference model, by a model with isometric knee extensor forces reduced to 40% of reference, by a model with isometric forces of all muscles reduced to 45% of reference, and by the reference model with the tibiofemoral compression force constrained during optimization to 65% of the peak value in the reference condition. INTERPRETATION: The strategies found by the model in conditions other than reference could be interpreted well on the basis of cost function and task biomechanics. The question remains whether it is feasible to teach patients with musculoskeletal impairments or joint pain to perform the sit-to-stand task according to strategies that are optimal according to the simulation model.

Optimizing the Distribution of Leg Muscles for Vertical Jumping.

A goal of biomechanics and motor control is to understand the design of the human musculoskeletal system. Here we investigated human functional morphology by making predictions about the muscle volume distribution that is optimal for a specific motor task. We examined a well-studied and relatively simple human movement, vertical jumping. We investigated how high a human could jump if muscle volume were optimized for jumping, and determined how the optimal parameters improve performance. We used a four-link inverted pendulum model of human vertical jumping actuated by Hill-type muscles, that well-approximates skilled human performance. We optimized muscle volume by allowing the cross-sectional area and muscle fiber optimum length to be changed for each muscle, while maintaining constant total muscle volume. We observed, perhaps surprisingly, that the reference model, based on human anthropometric data, is relatively good for vertical jumping; it achieves 90% of the jump height predicted by a model with muscles designed specifically for jumping. Alteration of cross-sectional areas-which determine the maximum force deliverable by the muscles-constitutes the majority of improvement to jump height. The optimal distribution results in large vastus, gastrocnemius and hamstrings muscles that deliver more work, while producing a kinematic pattern essentially identical to the reference model. Work output is increased by removing muscle from rectus femoris, which cannot do work on the skeleton given its moment arm at the hip and the joint excursions during push-off. The gluteus composes a disproportionate amount of muscle volume and jump height is improved by moving it to other muscles. This approach represents a way to test hypotheses about optimal human functional morphology. Future studies may extend this approach to address other morphological questions in ethological tasks such as locomotion, and feature other sets of parameters such as properties of the skeletal segments.

The cost of moving optimally: kinematic path selection.

It is currently unclear whether the brain plans movement kinematics explicitly or whether movement paths arise implicitly through optimization of a cost function that takes into account control and/or dynamic variables. Several cost functions are proposed in the literature that are very different in nature (e.g., control effort, torque change, and jerk), yet each can predict common movement characteristics. We set out to disentangle predictions of the different variables using a combination of modeling and empirical studies. Subjects performed goal-directed arm movements in a force field (FF) in combination with visual perturbations of seen hand position. This FF was designed to have distinct optimal movements for muscle-input and dynamic costs while leaving kinematic cost unchanged. Visual perturbations in turn changed the kinematic cost but left the dynamic and muscle-input costs unchanged. An optimally controlled, physiologically realistic arm model was used to predict movements under the various cost variables. Experimental results were not consistent with a cost function containing any of the control and dynamic costs investigated. Movement patterns of all experimental conditions were adequately predicted by a kinematic cost function comprising both visually and somatosensory perceived jerk. The present study provides clear behavioral evidence that the brain solves kinematic and mechanical redundancy in separate steps: in a first step, movement kinematics are planned; and in a second, separate step, muscle activation patterns are generated.

Bimanual proprioception: are two hands better than one?

Information about the position of an object that is held in both hands, such as a golf club or a tennis racquet, is transmitted to the human central nervous system from peripheral sensors in both left and right arms. How does the brain combine these two sources of information? Using a robot to move participant's passive limbs, we performed psychophysical estimates of proprioceptive function for each limb independently and again when subjects grasped the robot handle with both arms. We compared empirical estimates of bimanual proprioception to several models from the sensory integration literature: some that propose a combination of signals from the left and right arms (such as a Bayesian maximum-likelihood estimate), and some that propose using unimanual signals alone. Our results are consistent with the hypothesis that the nervous system both has knowledge of and uses the limb with the best proprioceptive acuity for bimanual proprioception. Surprisingly, a Bayesian model that postulates optimal combination of sensory signals could not predict empirically observed bimanual acuity. These findings suggest that while the central nervous system seems to have information about the relative sensory acuity of each limb, it uses this information in a rather rudimentary fashion, essentially ignoring information from the less reliable limb.

Humans make near-optimal adjustments of control to initial body configuration in vertical squat jumping.

We investigated adjustments of control to initial posture in squat jumping. Eleven male subjects jumped from three initial postures: preferred initial posture (PP), a posture in which the trunk was rotated 18° more backward (BP) and a posture in which it was rotated 15° more forward (FP) than in PP. Kinematics, ground reaction forces and electromyograms (EMG) were collected. EMG was rectified and smoothed to obtain smoothed rectified EMG (srEMG). Subjects showed adjustments in srEMG histories, most conspicuously a shift in srEMG-onset of rectus femoris (REC): from early in BP to late in FP. Jumps from the subjects' initial postures were simulated with a musculoskeletal model comprising four segments and six Hill-type muscles, which had muscle stimulation (STIM) over time as input. STIM of each muscle changed from initial to maximal at STIM-onset, and STIM-onsets were optimized using jump height as criterion. Optimal simulated jumps from BP, PP and FP were similar to jumps of the subjects. Optimal solutions primarily differed in STIM-onset of REC: from early in BP to late in FP. Because the subjects' adjustments in srEMG-onsets were similar to adjustments of the model's optimal STIM-onsets, it was concluded that the former were near-optimal. With the model we also showed that near-maximum jumps from BP, PP and FP could be achieved when STIM-onset of REC depended on initial hip joint angle and STIM-onsets of the other muscles were posture-independent. A control theory that relies on a mapping from initial posture to STIM-onsets seems a parsimonious alternative to theories relying on internal optimal control models.

Control of position and movement is simplified by combined muscle spindle and Golgi tendon organ feedback.

Whereas muscle spindles play a prominent role in current theories of human motor control, Golgi tendon organs (GTO) and their associated tendons are often neglected. This is surprising since there is ample evidence that both tendons and GTOs contribute importantly to neuromusculoskeletal dynamics. Using detailed musculoskeletal models, we provide evidence that simple feedback using muscle spindles alone results in very poor control of joint position and movement since muscle spindles cannot sense changes in tendon length that occur with changes in muscle force. We propose that a combination of spindle and GTO afferents can provide an estimate of muscle-tendon complex length, which can be effectively used for low-level feedback during both postural and movement tasks. The feasibility of the proposed scheme was tested using detailed musculoskeletal models of the human arm. Responses to transient and static perturbations were simulated using a 1-degree-of-freedom (DOF) model of the arm and showed that the combined feedback enabled the system to respond faster, reach steady state faster, and achieve smaller static position errors. Finally, we incorporated the proposed scheme in an optimally controlled 2-DOF model of the arm for fast point-to-point shoulder and elbow movements. Simulations showed that the proposed feedback could be easily incorporated in the optimal control framework without complicating the computation of the optimal control solution, yet greatly enhancing the system's response to perturbations. The theoretical analyses in this study might furthermore provide insight about the strong physiological couplings found between muscle spindle and GTO afferents in the human nervous system.

Can proprioceptive training improve motor learning?

Recent work has investigated the link between motor learning and sensory function in arm movement control. A number of findings are consistent with the idea that motor learning is associated with systematic changes to proprioception (Haith A, Jackson C, Mial R, Vijayakumar S. Adv Neural Inf Process Syst 21: 593-600, 2008; Ostry DJ, Darainy M, Mattar AA, Wong J, Gribble PL. J Neurosci 30: 5384-5393, 2010; Vahdat S, Darainy M, Milner TE, Ostry DJ. J Neurosci 31: 16907-16915, 2011). Here, we tested whether motor learning could be improved by providing subjects with proprioceptive training on a desired hand trajectory. Subjects were instructed to reproduce both the time-varying position and velocity of novel, complex hand trajectories. Subjects underwent 3 days of training with 90 movement trials per day. Active movement trials were interleaved with demonstration trials. For control subjects, these interleaved demonstration trials consisted of visual demonstration alone. A second group of subjects received visual and proprioceptive demonstration simultaneously; this group was presented with the same visual stimulus, but, in addition, their limb was moved through the target trajectory by a robot using servo control. Subjects who experienced the additional proprioceptive demonstration of the desired trajectory showed greater improvements during training movements than control subjects who only received visual information. This benefit of adding proprioceptive training was seen in both movement speed and position error. Interestingly, additional control subjects who received proprioceptive guidance while actively moving their arm during demonstration trials did not show the same improvement in positional accuracy. These findings support the idea that the addition of proprioceptive training can augment motor learning, and that this benefit is greatest when the subject passively experiences the goal movement.

In vivo dynamics of the musculoskeletal system cannot be adequately described using a stiffness-damping-inertia model.

BACKGROUND: Visco-elastic properties of the (neuro-)musculoskeletal system play a fundamental role in the control of posture and movement. Often, these properties are described and identified using stiffness-damping-inertia (KBI) models. In such an approach, perturbations are applied to the (neuro-)musculoskeletal system and subsequently KBI-model parameters are optimized to obtain a best fit between simulated and experimentally observed responses. Problems with this approach may arise because a KBI-model neglects critical aspects of the real musculoskeletal system. METHODOLOGY/PRINCIPAL FINDINGS: The purpose of this study was to analyze the relation between the musculoskeletal properties and the stiffness and damping estimated using a KBI-model, to analyze how this relation is affected by the nature of the perturbation and to assess the sensitivity of the estimated stiffness and damping to measurement errors. Our analyses show that the estimated stiffness and damping using KBI-models do not resemble any of the dynamical parameters of the underlying system, not even when the responses are very accurately fitted by the KBI-model. Furthermore, the stiffness and damping depend non-linearly on all the dynamical parameters of the underlying system, influenced by the nature of the perturbation and the time interval over which the KBI-model is optimized. Moreover, our analyses predict a very high sensitivity of estimated parameters to measurement errors. CONCLUSIONS/SIGNIFICANCE: The results of this study suggest that the usage of stiffness-damping-inertia models to investigate the dynamical properties of the musculoskeletal system under control by the CNS should be reconsidered.

The central nervous system does not minimize energy cost in arm movements.

It has been widely suggested that the many degrees of freedom of the musculoskeletal system may be exploited by the CNS to minimize energy cost. We tested this idea by having subjects making point-to-point movements while grasping a robotic manipulandum. The robot created a force field chosen such that the minimal energy hand path for reaching movements differed substantially from those observed in a null field. The results show that after extended exposure to the force field, subjects continued to move exactly as they did in the null field and thus used substantially more energy than needed. Even after practicing to move along the minimal energy path, subjects did not adapt their freely chosen hand paths to reduce energy expenditure. The results of this study indicate that for point-to-point arm movements minimization of energy cost is not a dominant factor that influences how the CNS arrives at kinematics and associated muscle activation patterns.

Equilibrium point control cannot be refuted by experimental reconstruction of equilibrium point trajectories.

In the literature, it has been hotly debated whether the brain uses internal models or equilibrium point (EP) control to generate arm movements. EP control involves specification of EP trajectories, time series of arm configurations in which internal forces and external forces are in equilibrium; if the arm is not in a specified EP, it is driven toward this EP by muscle forces arising due to central drive, reflexes, and muscle mechanics. EP control has been refuted by researchers claiming that EP trajectories underlying movements of subjects were complex. These researchers used an approach that involves applying force perturbations during movements of subjects and fitting a mass-spring-damper model to the kinematic responses, and then reconstructing the EP trajectory using the estimated stiffness, damping, and measured kinematics. In this study, we examined the validity of this approach using an EP-controlled musculoskeletal model of the arm. We used the latter model to simulate unperturbed and perturbed maximally fast movements and optimized the parameter values of a mass-spring-damper model to make it reproduce as best as possible the kinematic responses. It was shown that estimated stiffness not only depended on the "true" stiffness of the musculoskeletal model but on all of its dynamical parameters. Furthermore it was shown that reconstructed EP trajectories were in agreement with those presented in the literature but did not resemble the simple EP trajectories that had been used to generate the movement of the model. It was concluded that the refutation of EP control on the basis of results obtained with mass-spring-damper models was unjust.

A model of open-loop control of equilibrium position and stiffness of the human elbow joint.

According to the equilibrium point theory, the control of posture and movement involves the setting of equilibrium joint positions (EP) and the independent modulation of stiffness. One model of EP control, the alpha-model, posits that stable EPs and stiffness are set open-loop, i.e. without the aid of feedback. The purpose of the present study was to explore for the elbow joint the range over which stable EPs can be set open-loop and to investigate the effect of co-contraction on intrinsic low-frequency elbow joint stiffness (K (ilf)). For this purpose, a model of the upper and lower arm was constructed, equipped with Hill-type muscles. At a constant neural input, the isometric force of the contractile element of the muscles depended on both the myofilamentary overlap and the effect of sarcomere length on the sensitivity of myofilaments to [Ca2+] (LDCS). The musculoskeletal model, for which the parameters were chosen carefully on the basis of physiological literature, captured the salient isometric properties of the muscles spanning the elbow joint. It was found that stable open-loop EPs could be achieved over the whole range of motion of the elbow joint and that K (ilf), which ranged from 18 to 42 N m.rad(-1), could be independently controlled. In the model, LDCS contributed substantially to K (ilf) (up to 25 N m.rad(-1)) and caused K (ilf) to peak at a sub-maximal level of co-contraction.

Is equilibrium point control feasible for fast goal-directed single-joint movements?

Several types of equilibrium point (EP) controllers have been proposed for the control of posture and movement. EP controllers are appealing from a computational perspective because they do not require solving the "inverse dynamic problem" (i.e., computation of the torques required to move a system along a desired trajectory). It has been argued that EP controllers are not capable of controlling fast single-joint movements. To refute this statement, several extensions have been proposed, although these have been tested using models in which only the tendon compliance, force-length-velocity relation, and mechanical interaction between tendon and contractile element were not adequately represented. In the present study, fast elbow-joint movements were measured and an attempt was made to reproduce these using a realistic musculoskeletal model of the human arm. Three types of EP controllers were evaluated: an open-loop alpha-controller, a closed-loop lambda-controller, and a hybrid open- and closed-loop controller. For each controller we considered a continuous version and a version in which the control signals were sent out intermittently. Only the intermittent hybrid EP controller was capable of generating movements that were as fast as those of the subjects. As a result of the nonlinear muscle properties, the hybrid EP controller requires a more detailed representation of static muscle properties than generally assumed in the context of EP control. In sum, this study shows that fast single-joint movements can be realized without explicitly solving the inverse dynamics problem, but in a less straightforward manner than implied by proponents of conventional EP controllers.

Length-dependent [Ca2+] sensitivity adds stiffness to muscle.

It is well documented that muscle fibers become more sensitive for [Ca2+] with increasing sarcomere length. In mechanical terms this length-dependent [Ca2+] sensitivity (LDCS) adds to the stiffness of muscle fibers, because muscle force, normalized for the force-length relationship at maximal stimulation, increases with contractile element (CE) length. Although LDCS is well-documented in the physiological literature, it is ignored in most motor control studies. The aim of the present study was to investigate the importance of LDCS as a contributor to the stiffness of a muscle. Comparison of experimental data with predictions derived from the model of activation dynamics proposed by Hatze (Myocybernetic Control Models of Skeletal Muscle, University of South Africa, Pretoria, 1981, pp. 31-42) indicated that this model captures the main characteristics of LDCS well. It was shown that LDCS accounts for the experimentally observed shifts in optimum length at sub-maximal stimulation levels. Furthermore, it was shown that in conditions with low-to-medium muscle stimulation, the contribution of LDCS to the total amount of stiffness provided by the muscle is substantial. It was concluded that LDCS is an important muscle property and should be taken into account in studies concerning motor control.