Lukas Jan Volz

Adaptive behaviour depends on continuously updating motor control when action outcomes differ from expectations. Because action execution and sensory feedback are noisy, the value of an error depends on how reliably it reflects the consequences of an action. Most studies have manipulated the reliability of sensory feedback or action outcomes rather than uncertainty generated by the action itself. Here we show that motor adaptation is calibrated to effort-dependent uncertainty arising during action generation. Using a force-based motor adaptation paradigm, we demonstrate that greater effort progressively increased uncertainty in action execution and proprioceptive feedback, with corresponding changes in adaptation rate that closely matched normative predictions. Reward enhanced adaptation by improving proprioceptive precision rather than directly changing adaptation rate. Supporting this framework, patients with major depressive disorder exhibited elevated execution and proprioceptive uncertainty, impaired reward-dependent improvement in proprioceptive precision, and slower motor adaptation. These findings establish intrinsic uncertainty as a general determinant of adaptive behaviour, linking action generation, sensory evaluation, and motivation.

Background: Symptom-based localization of brainstem ischemia is challenging because of the anatomical complexity of the brainstem and the nonspecific overlap of clinical syndromes. Whether large language models (LLMs) can meaningfully assist in this task remains uncertain. Objective: This study aimed to compare the performance of several OpenAI LLMs and neurologists in localizing diffusion-weighted imaging (DWI)–confirmed brainstem ischemic lesions based on symptom descriptions alone. Methods: In this retrospective single-center study, 109 patients with DWI-confirmed acute brainstem ischemia were included. Three neurologists and 6 LLMs (GPT-5, GPT-4, GPT-4.1, GPT-4o, o3, and o3-pro) predicted lesion localization using a combined anatomical-lateral end point (left or right midbrain, pons, and medulla) based on symptom descriptions alone. Overall and regional accuracy, the Cohen κ, 6-class confusion matrices, and point-biserial correlations between symptom count and correct prediction were assessed. Because all raters evaluated the same cases, paired McNemar tests with Benjamini-Hochberg correction were used for pairwise performance comparisons. Results: GPT-4 and GPT-4o achieved the highest overall accuracy (61/109, 56%; 95% CI 46.1%‐65.5%). Agreement with the DWI reference standard remained limited across all raters, with the Cohen κ reaching a maximum of 0.291 for GPT-4o. Confusion matrices showed that higher performance was driven mainly by pontine cases, whereas misclassification remained frequent in mesencephalic and medullary lesions. Regional analyses outside the pons were imprecise because mesencephalic and medullary subgroups each contained only 16 cases. A higher number of documented symptoms was associated with correct prediction for GPT-4, GPT-5, GPT-o3, and 1 neurologist. Conclusions: Although some LLMs showed higher relative accuracy than the participating neurologists, absolute performance remained limited and clinically insufficient. These findings are best interpreted as an exploratory benchmark under constrained conditions: absolute performance remained modest, agreement beyond chance was limited, and performance outside pontine lesions was inconclusive.

Background: Corticospinal tract (CST) damage is a major cause of post-stroke motor deficits. However, it remains unclear which estimates of CST damage best predict motor recovery, especially regarding different aspects of motor control. While conventional CST-lesion metrics offer superior feasibility, data-driven machine learning (ML) approaches may better capture patients’ propensity for task-specific recovery with important implication for their use as future clinical biomarkers. Methods: Providing the first direct longitudinal comparison of these approaches based exclusively on CST-lesion patterns, we evaluated six conventional CST-lesion metrics and a voxel-wise ML approach using clinical MRI data from 127 acute ischemic stroke patients. Acute impairment and outcome (>3 months post-stroke) were assessed for basal and complex motor functions. Conventional CST-lesion metrics and ML were used to predict task-specific motor impairment and outcome. Results: All conventional CST-lesion metrics correlated significantly with both acute impairment and motor outcome across motor domains, with metrics weighted for CST narrowing and tract probability performing best. However, predictive performance for unseen patients was low. ML outperformed conventional markers in predicting acute impairment across motor domains and basal motor outcome, but failed to predict complex motor outcome. Topographically, predictive voxels clustered within and above the posterior limb of the internal capsule, with distinct CST subregions associated with basal versus complex motor impairment, consistent with a task-specific somatotopic organization. Conclusions: The predictive utility of CST biomarkers was task- and timepoint-dependent. While ML may improve predictive performance, complex motor outcome remained difficult to predict, likely reflecting greater reliance on distributed cortical reorganization beyond the CST. By revealing task-specific CST subregions, voxel-wise ML provides an anatomically informed foundation for future predictive models. Such future models should combine CST biomarkers with measures of broader motor network integrity to enable individualized prognosis tailored to specific motor domains and recovery stages.

While reinforcement has repeatedly been shown to impact motor adaptation, previous inconsistent findings hinder our mechanistic understanding. To conciliate contradictory views, we used two distinct motor tasks to investigate whether reinforcement differentially modulates two core motor learning processes: between-trial learning resulting from motor planning and within-trial learning supported by the online correction of ongoing movements. We observed successful adaptation, retention and highly similar temporal profiles of adaptation for both tasks, in line with the notion that motor adaptation of upper limb movements is likely to result from updates in both motor planning and online correction. Reinforcement enhanced performance in the task allowing for within-trial adaptation. Conversely, higher degrees of early motor variability were associated with faster between-trial but not within-trial adaptation. These results highlight a dissociation of factors affecting motor adaptation: Reinforcement seems to primarily impact within-trial learning linked to the online correction of ongoing movements. In contrast, early motor variability might benefit between-trial learning by informing motor planning potentially via an exploration of the motor output space. Our results help explain previous inconsistent findings and add to our theoretical understanding by demonstrating that the effects of reinforcement and motor variability may depend on whether adaptation is driven by within - or between-trial learning.

Structural cortico-cortical connectivity is essential for motor recovery after stroke, with undamaged fibre tracts potentially serving as structural reserve for functional reorganization. Yet, the mechanisms by which the structural reserve contributes to changes in functional network configurations to improve post-stroke motor control remain unknown. Here, we assessed structural (diffusion spectrum imaging) and effective (functional MRI-based dynamic causal modelling) connectivity to examine how the structural reserve may guide motor network reorganization of upper limb motor control. Specific features of cortico-cortical structural reserve were associated with distinct patterns of functional network configurations: limited intrahemispheric structural reserve between ipsilesional primary motor cortex and premotor areas was linked to interhemispheric rerouting of motor commands via the contralesional primary motor cortex, particularly when ipsilesional corticospinal tract integrity was low. In contrast, limited interhemispheric structural reserve between ipsilesional primary motor cortex and contralesional premotor areas was related to intrahemispheric rerouting via the ipsilesional premotor cortex, especially in patients with high residual corticospinal tract integrity. Even though both alternative pathways may allow bypassing damaged corticospinal tract fibres originating from the ipsilesional primary motor cortex, we observed a clear behavioural dissociation: in patients who substantially recovered, enhanced intrahemispheric rerouting was indicative of better hand function, reflecting beneficial reorganization. Conversely, interhemispheric rerouting was associated with pronounced motor impairment of the paretic arm and primarily observed in non-substantially recovered patients, in line with task-specific maladaptive reorganization. Our findings emphasize that the motor network’s available structural reserve critically shapes functional network reorganization by predetermining whether motor commands are primarily rerouted intra- or interhemispherically with distinct implications for motor recovery. Besides helping to reconcile previous conflicting interpretations on the role of contralesional primary motor cortex, our results underscore that the individual level of structural motor network reserve should be considered for future therapeutic approaches aiming at amplifying motor recovery after stroke.

The dynamic integration of the lateralized and specialized capacities of the two cerebral hemispheres constitutes a hallmark feature of human brain function. This interhemispheric exchange of information critically depends upon the corpus callosum. Classical anatomical descriptions of callosal organization outline a topographic gradient from front to back, such that specific transcallosal fibers support distinct aspects of integrated brain function. Here, we present a challenge to this conventional model. Using neuroimaging data obtained from a new cohort of adult corpus callosotomy patients, we leverage modern network neuroscience techniques to show that full interhemispheric integration can be achieved via a small proportion of posterior callosal fibers. Partial callosotomy patients with spared callosal fibers retained widespread patterns of interhemispheric functional connectivity and showed no signs of behavioral disconnection, even with only 1 cm of the splenium intact. Conversely, only complete callosotomy patients demonstrated sweeping disruptions of interhemispheric network architectures, aligning with disconnection syndromes long-thought to reflect diminished information propagation and communication across the brain. These findings motivate an evolving mechanistic understanding of synchronized interhemispheric neural activity for large-scale human brain function and behavior.

BACKGROUND: Stroke is increasingly understood as a network disorder with symptoms often arising from disruption of white matter connectivity. Previous connectome-based lesion-symptom mapping studies revealed that poststroke motor deficits are not only associated with damage to the core sensorimotor network but also with nonsensorimotor connections. However, whether task-specific initial impairment and outcome are based on distinct disconnection patterns remains unknown. METHODS: To address this question, we included lesion information and assessments of distinct aspects of upper limb motor impairment of 113 patients with early subacute stroke (mean age, 65.95 years). We used connectome-based lesion-symptom mapping, based on a normative structural connectome, and a machine learning algorithm to predict individual levels of task-specific motor impairment and outcome >3 months later. RESULTS: We identified task-specific disconnection patterns that significantly predicted initial motor impairment and outcome and a task-general reach-to-grasp network including both sensorimotor and nonsensorimotor areas. More complex reach-to-grasp movements showed a substantial overlap in disconnections for the prediction of impairment and outcome. Conversely, disconnections indicative of more basal aspects of motor control substantially differed between the prediction of initial impairment and outcome at the chronic stage poststroke. Similarly, the significance of interhemispheric disconnections changed in a task- and time-dependent fashion. CONCLUSIONS: In summary, our study identified distinct disconnection patterns indicative of specific aspects of motor impairment and outcome after stroke, highlighting a time- and task-dependent role of the contralesional hemisphere and suggesting a domain-general compensatory role of nonsensorimotor temporal areas. From a mechanistic perspective, differences in disconnection patterns predictive of initial motor impairment versus outcome suggest a stronger dependence of basal motor control on the brain’s structural reserve during motor recovery. Our results extend our current network-level understanding of task-specific motor impairment and recovery, and emphasize the potential of connectome-based lesion-symptom mapping for future clinical applications.

Tic disorders (TD) are childhood-onset neurodevelopmental disorders characterized by sudden, repetitive motor and vocal tics, often with partial or complete remission by the time young adulthood is reached. We here investigated motor control and compensatory neural processes in drug-naïve children and adolescents with chronic Motor Tic Disorder or Tourette Syndrome (TD) by examining motor network activity and connectivity compared to healthy controls. Using a reaction time (RT) task under varying cueing conditions, combined with functional magnetic resonance imaging (fMRI) and dynamic causal modelling (DCM), we explored how TD-related motor networks adapt to support volitional movement control. Participants with TD demonstrated enhanced task accuracy across internally and externally cued conditions despite deficits in sustained motor inhibition (blink suppression). Relative to controls, individuals with TD exhibited increased task-related activation in ipsilateral motor regions, particularly in the primary motor cortex, and somatosensory cortex, and enhanced interhemispheric connectivity between parietal sensory-motor hubs. Notably, while in typically developing participants, age-related increases in parietal lobe activation and modulatory connectivity between primary motor and premotor regions were linked to improved task accuracy, working memory and visuomotor coordination, TD patients deviated from this normative developmental trajectory with distinct, atypical but neither delayed nor accelerated neural activation and connectivity patterns. Our data suggest that TD involves compensatory neuroplastic adaptations that leverage additional sensorimotor resources to improve motor control but do not extend to motor inhibition processes. Moreover, the findings emphasize the intricate interplay between motor control and neural plasticity in TD, highlighting how compensatory mechanisms may serve as adaptive responses to motor challenges. These findings open avenues for therapeutic strategies that harness the brain's compensatory capacities to enhance motor control and facilitate TD management.

Sensorimotor processing in the human brain is largely lateralized, with the corpus callosum integrating these processes into a unified experience. Following complete callosotomy, this integration breaks down, resulting in disconnection syndromes. We asked how much of the corpus callosum is sufficient to support functional unity—the absence of disconnection syndrome—by comparing three complete callosotomy patients with one retaining only the splenium. Using lateralized tasks across visual, tactile, visuospatial, and language domains, we predicted domain-specific deficits in the splenium-only patient based on established anatomical models of callosal topography. Strikingly, while complete callosotomy patients exhibited disconnection syndromes, the splenium patient demonstrated functional unity across all domains—as if his entire corpus callosum were intact. Our findings highlight the brain’s remarkable capacity to maintain behavioral integration through minimal preserved pathways, highlighting how the structure-dependent reorganizational capacity of the human brain allows to preserve functional unity.

The dynamic integration of the lateralized and specialized capacities of the cerebral hemispheres constitutes a hallmark feature of human brain function. This inter-hemispheric exchange of information critically depends upon the corpus callosum. Classical descriptions of callosal organization outline a topographic gradient, such that specific fibers integrate distinct aspects of brain function. Here we present a challenge to this conventional model. Using neuroimaging data obtained from a new cohort of adult corpus callosotomy patients, we leverage modern network neuroscience techniques to show - for the first time - that full inter-hemispheric integration can be achieved via a small proportion (~1 cm) of intact posterior callosal fibers. Only complete callosotomy patients demonstrated the expected dissolution of typical inter-hemispheric network architectures, aligning with disconnection syndromes long-thought to reflect diminished information propagation and communication across the brain. These findings motivate a novel mechanistic understanding of synchronized inter-hemispheric neural activity for large-scale human brain function and behavior.