The research team, led by Vishal Rawji and Cosima Graef of Imperial College London and King’s College London, with senior supervision from Mark Edwards and Dario Farina, recruited twelve people with Parkinson’s disease who were treated with subthalamic deep brain stimulation, alongside ten age-matched healthy controls. Deep brain stimulation was a critical design feature. Because the therapy temporarily suppresses the cardinal motor symptoms of the disease, switching it on and off allowed the investigators to observe the same patients’ neuromuscular systems in both a medicated-device state and a more symptomatic state, effectively doubling the contrast available from a modest sample.

The technical heart of the study was high-density surface electromyography, a non-invasive technique that records electrical activity from dozens of closely spaced electrodes placed over the skin. Modern decomposition algorithms can separate this composite signal into the firing patterns of individual motor units, the spinal motor neurons and their associated muscle fibres that represent the final common pathway of all voluntary movement. By tracking these units during precisely controlled force tasks, the researchers could interrogate, at the level of single neurons, whether the nervous system of a person with Parkinson’s can still assemble and deliver a properly sized burst of drive to muscle.

Participants performed ballistic isometric contractions, rapid force efforts against a fixed resistance, at three target levels: ten, thirty and fifty percent of their maximum voluntary force. In healthy controls, these movements were typically accomplished with a single compact burst of motor unit activity, averaging 1.41 bursts per contraction. Patients with stimulation on behaved differently, recruiting 2.57 bursts on average, and when their stimulation was switched off the number rose further to 2.90 bursts, coinciding with a measurable worsening of bradykinesia. Instead of one decisive command, the patients’ motor systems issued a series of smaller, successive pulses.

The most striking finding emerged when the researchers examined the size of the patients’ first bursts during the harder contractions. At the thirty and fifty percent force targets, the initial burst of motor unit activity was large enough to have achieved the ten and thirty percent targets respectively. In other words, the patients demonstrably could generate neural drives of the appropriate magnitude for easier movements; they simply did not deploy them at the correct scale for the task at hand. This dissociation between capacity and deployment is precisely what the motor vigour hypothesis predicts, and it is difficult to reconcile with an account in which the drive-generating machinery itself is broken.

Further evidence came from the detailed properties of motor unit firing. During both the ballistic contractions and separate repetitive contraction tasks, the discharge rate of motor units and its variability did not differ between patients and controls, nor between stimulation-on and stimulation-off conditions. If the mechanisms for generating neural drive, or for transmitting it from spinal motor neurons to muscle fibres, were impaired in Parkinson’s, one would expect these parameters to shift with disease state and with the suppression of symptoms by stimulation. Their stability across all conditions indicates a preserved low-level neuromuscular apparatus.

What, then, explains the slowness? The answer lies in timing. Because the patients’ movements were assembled from multiple successive bursts separated by delays, they reached the target force roughly 230 milliseconds later than controls. In everyday terms, the movement is not weak so much as hesitant: the nervous system issues an underscaled command, registers the shortfall, and issues another, repeating the process until the target is met. Each extra step in that iterative chain adds latency, and the accumulated cost is the characteristic bradykinesia and hypokinesia, the reduced amplitude and speed of movement, that clinicians observe at the bedside.

The findings carry conceptual weight for how the field understands basal ganglia function. The subthalamic nucleus and its neighbours are increasingly viewed not as simple relays but as modulators of movement vigour, the internal gain setting that determines how boldly a planned action is executed. The present results align with this framework: with stimulation restoring more appropriate scaling, patients converge toward the single-burst strategy of controls, while the raw capacity for drive generation remains untouched by the disease process. Slowness, on this reading, is a calibration problem rather than a transmission failure.

There are also practical implications. Rehabilitation and behavioural therapies for Parkinson’s have long exploited external cues and reinforcement to encourage larger, faster movements, approaches that make intuitive sense if the underlying system is intact but miscalibrated. The study’s methodology, combining high-density surface EMG with single-motor-unit decomposition in patients with implanted devices, offers a template for quantifying how such interventions reshape the scaling of motor commands. It may also inform the programming of deep brain stimulation itself, since the on-off contrast showed that symptom severity tracked the fragmentation of motor unit activity in a graded way.

Caveats remain. The patient cohort was small and specifically composed of individuals with subthalamic implants, so the findings may not generalise uniformly to all people with Parkinson’s or to those managed with medication alone. The tasks examined were isometric and constrained, whereas natural movement unfolds across many joints with sensory feedback continuously in play. Nevertheless, by demonstrating that patients can generate appropriately sized neural drives but fail to scale them, the study provides some of the most direct single-neuron evidence to date that the slowness of Parkinson’s disease reflects a vigour deficit in motor command generation, not a degraded capacity to drive muscle. The distinction matters, because it redirects attention from the muscle and the final motor pathway toward the higher-order decision of how much force a movement deserves.

Subject of Research: Neural drive generation and motor vigour in Parkinson's disease bradykinesia

Article Title: Hypokinesia and bradykinesia in Parkinson’s disease are consistent with reduced motor vigour, not an impaired capacity to generate and transmit neural drive to muscle

Article References: Rawji, V., Graef, C., Foltynie, T., Haar, S., Tai, Y. F., Edwards, M., & Farina, D. (2026). Hypokinesia and bradykinesia in Parkinson’s disease are consistent with reduced motor vigour, not an impaired capacity to generate and transmit neural drive to muscle. npj Parkinson's Disease. https://doi.org/10.1038/s41531-026-01590-7

Image Credits: AI Generated

DOI: 10.1038/s41531-026-01590-7

Keywords: Parkinson's disease, bradykinesia, hypokinesia, motor vigour, neural drive, high-density surface EMG, motor units, deep brain stimulation, subthalamic nucleus, ballistic contractions, basal ganglia, motor control

The research team, led by investigators at Deakin University in Australia, recruited fifty-six older adults spanning three distinct ageing phenotypes: eleven people with clinically diagnosed sarcopenia, twenty-two non-sarcopenic controls, and nineteen masters athletes with at least ten years of continuous competitive sport participation. Sarcopenia was diagnosed according to the Sarcopenia Definitions and Outcomes Consortium criteria, which combine low handgrip strength with slow usual gait speed. The sarcopenic participants were, on average, the oldest of the three groups at just over eighty years, while the masters athletes were the youngest at around seventy. Crucially, the researchers went to considerable lengths to ensure that age itself did not explain their findings, running sensitivity analyses with age adjustment and an exploratory age-matched comparison that preserved the central results.

The experimental task was elegantly simple in design but technically ambitious in execution. Seated with their preferred leg strapped into an ankle dynamometer, participants performed trapezoidal isometric contractions of the tibialis anterior, the shin muscle that lifts the foot and clears the toes during walking. Each contraction required ramping up to thirty percent of maximal voluntary torque, holding that plateau for twenty seconds, and ramping down again. The tibialis anterior was chosen deliberately: its control is directly relevant to foot clearance during gait, and impaired control of this muscle is linked to trips and falls. Two sixty-four-channel high-density electrode grids were placed over the muscle belly, allowing the researchers to decompose the surface EMG signal into the firing times of individual motor units, the final common pathway through which the nervous system commands muscle.

The dual-task manipulation added a cognitive burden to this motor challenge. While holding the steady contraction, participants were asked to continuously subtract seven from a starting number between seventy and ninety-nine, a classic serial subtraction task used to tax executive function. Trials were invalidated if the subtraction was interrupted for more than three seconds or if multiple errors were made, ensuring that participants genuinely engaged with both tasks. Single-task and dual-task trials were alternated in randomised order to distribute any residual fatigue evenly across conditions. Real-time visual torque feedback was displayed on a monitor with identical settings for every participant, and plateau torque remained close to the thirty percent target in all groups under both conditions, meaning that any differences in motor unit behaviour could not be attributed to differences in the force actually produced.

The headline finding was stark. Torque steadiness, quantified as the coefficient of variation of the force signal, was already poorer in the sarcopenic group during single-task contractions compared with controls and athletes. When the cognitive task was added, steadiness deteriorated further in the sarcopenic participants alone, while controls and masters athletes held their performance stable. In total, the team identified more than 8,300 motor unit spike trains across the cohort, providing an unusually rich dataset. The mean discharge rate of motor units increased similarly in all three groups during dual-tasking, rising by only about 2.6 percent, which told the researchers that the sarcopenic deterioration was not about firing harder or faster. Instead, the problem lay in the timing: the regularity of motor unit discharge, measured as the variability of inter-spike intervals, worsened markedly in the sarcopenic group under cognitive load, while it actually improved in the masters athletes and remained unchanged in controls.

To understand where in the nervous system these differences arose, the researchers turned to intramuscular coherence analysis, a technique that estimates the common synaptic input shared across the motoneuron pool. By summing the spike trains of randomly selected groups of five motor units into cumulative spike trains and computing coherence spectra across up to one hundred permutations, they quantified how strongly different motor units were driven together at different frequencies. Three frequency bands were examined separately, each thought to reflect a different source of shared neural input. Delta-band coherence, spanning one to five hertz, is associated with the low-frequency common drive linked to force fluctuations. Alpha-band coherence, from five to fifteen hertz, is associated with physiological tremor and spinal and afferent contributions. Beta-band coherence, from fifteen to thirty-five hertz, is widely linked to corticospinal coupling, the dialogue between cortex and spinal motor neurons.

The coherence results painted a nuanced picture. Delta-band coherence decreased in all three groups during dual-tasking, suggesting a broad redistribution of low-frequency common input whenever attention was divided, but because this occurred even in the steady performers, it could not explain the sarcopenia-specific loss of force control. The decisive divergence appeared in the alpha and beta bands. Only the sarcopenic group showed an increase in alpha-band coherence under cognitive load, a rise in oscillatory common input that the authors interpret as a form of neural noise, tremor-related activity that is expressed in the force signal and degrades the accuracy of force production. This increase coincided with the greater inter-spike interval variability and torque variability observed in the same participants, although the study design does not establish a causal pathway between these measures. Meanwhile, beta-band coherence rose in the sarcopenic group but fell in controls and, most prominently, in masters athletes, hinting at fundamentally different strategies for regulating cortical input when attention is stretched.

The masters athletes emerged as the most striking contrast in the study. They were the only group whose discharge regularity improved during dual-tasking, and they showed the largest reduction in beta-band coherence, consistent with a flexible, adaptable modulation of common synaptic input. The authors suggest that decades of training in dynamic environments, which demand continuous integration of sensory information, attentional shifting, decision-making and motor execution, may promote more efficient allocation of central resources and more resilient control strategies. Importantly, exploratory comparisons between endurance-type and power-type athletes showed broadly similar dual-task responses, though the small subgroup sizes make those comparisons tentative. The broader message is that neural adaptability in the ageing motor system appears modifiable, and lifelong varied training may be one route to preserving it.

The clinical implications reach well beyond the laboratory. Force steadiness has been linked to mobility performance and postural control, and dual-task ability is strongly connected to fall risk and loss of independence in older adults. In an exploratory analysis drawing on the team’s companion study, a larger dual-task increase in force variability was associated with poorer baseline functional power and mobility, but not with maximal dorsiflexion strength, suggesting that vulnerability to cognitive loading reflects broad functional capacity rather than raw muscle strength. This implies that conventional single-task strength testing may underestimate real-world risk in people with sarcopenia, precisely because everyday activities such as walking, rising from a chair and carrying objects are performed while attention is divided. The findings support interventions that go beyond resistance training alone, incorporating balance-challenging, multicomponent and dual-task exercise programmes that train the nervous system as well as the muscle, and they add momentum to a reframing of sarcopenia as a disease of the ageing motor system rather than of muscle mass alone.

Subject of Research: Neural control of force steadiness during dual-task contractions in older adults with sarcopenia

Article Title: Dual‐Tasking Exacerbates Force and Neural Control Unsteadiness in Older Adults With Sarcopenia

Article References: Orssatto, L. B. R., Clark, B. C., Scott, D., Cabral, H. V., Fernandes, G. L., & Daly, R. M. (2026). Dual‐Tasking Exacerbates Force and Neural Control Unsteadiness in Older Adults With Sarcopenia. Journal of Cachexia, Sarcopenia and Muscle, 17(5), Article e70397. https://doi.org/10.1002/jcsm.70397

Image Credits: AI Generated

DOI: 10.1002/jcsm.70397

Keywords: sarcopenia, motor units, force steadiness, dual-task, high-density EMG, coherence, ageing, masters athletes, neural drive, falls, cognitive load, neuromuscular control