UPPER LIMB MOTOR CONTROL IN PEOPLE WITH STROKE: EXPLORING SENSORY AND MOTOR COMPONENTS OF FORWARD MODELS DYSFUNCTIONS

A crucial feature of the upper limb motor control is the possibility of predicting the consequences of our own movements. The literature suggested that the central nervous system uses forward models that, starting from the initial position of the upper limb and from the action’s motor command, predict the sensory feedback and the future limb’s state during the action execution. The cerebellum is considered a crucial structure for the neural implementation of such internal model. This evidence is supported by a series of studies that described the cerebellar ataxic behavior as a forward model dysfunction. However, less attention has been dedicated to the role of somatosensory information in predictive motor control for the upper limbs. The goal of this project was to investigate different aspects of the anticipatory motor control in healthy subjects and in people with cerebral stroke with cerebellar lesions and somatosensory impairment. In the first section, we aimed to provide further evidence on the role of the forward model in upper limbs sensorimotor function. We focused on the sensory attenuation phenomenon, and we performed new analyses on previously recorded data on force matching tasks in healthy subjects. We collected data from three experiments including a total of 375 subjects. In these experiments a target force was delivered on the subjects’ left index finger and they asked to reproduce it in different conditions. Our analysis compared the within-subject trial-to-trial variability in the matching force between conditions. The results showed that, considering the matching force exertion, the trial-to-trial variability was associated with the level of tactile perception, corroborating the hypothesis that a predictive mechanism is involved in the sensory attenuation phenomenon. In the second section, we aimed to investigate anticipatory grip force modulation deficits during a bimanual object lifting task in patients with stroke and somatosensory deficits. The assessment procedures included two force matching tasks and a lifting task. We used a sensorized device allowing us to record the grip force with which subjects grasped it and track its position and compute kinematics parameters of the lifting. After validating our experimental procedures in healthy controls, we included 11 healthy subjects and 9 patients with stroke and somatosensory deficits. Our analysis provided some evidence of impaired motor planning in patients with CNS stroke sequelae. Specifically, patients with stroke showed abnormal timing of maximal grip force exertion during the lifting. In the last section, we aimed to assess predictive upper limb behavior of a fast and repetitive reaching task in patients with cerebellar lesions and patients with somatosensory deficits. We developed and validated an accurate low-cost system for the kinematic assessment of the index-to-nose task and we compared a group of young healthy subjects, and stroke patients with cerebellar lesions or proprioceptive impairment. Our analysis measured a set of reaching parameters referring to movement accuracy, efficiency and motor planning in a cohort of healthy controls and patients affected by CNS stroke. Moreover, we proposed a theoretical framework for interpreting sensory and cerebellar ataxia as different forward model’s dysfunctions. Our results suggested that patients with somatosensory deficit were the most affected by the absence of visual feedback significantly reducing movement speed when performing the task with closed eyes. Moreover, we found that patients with cerebellar lesions showed signs of impaired movement planning. This dissertation underlined the importance of assessing predictive motor control in patients affected by CNS lesions. Future research should focus on the role of anticipatory motor control in motor learning and on the design of rehabilitation treatment for forward model dysfunctions.