Clinical assessment of motor and sensory deficits is still today largely based on tests that do not permit any precise quantification. However, robotic technologies, coupled with neuroimaging techniques constitute new tools to assess sensorimotor functions that could allow to conceive neurorehabilitation protocols better adapted to the neurological impairment of each patient and to her/his specific recovery profile.

The goal of this project is to contribute identifying the factors that determine functional recovery in stroke patients presenting upper-limb motor deficits. Here, we will focus our research on two factors that contribute in a complementary way to motor control: 1) the processing of proprioceptive informations, and 2) the processing of movement-execution errors. In this purpose, we will combine psychophysical methods that allow to precisely quantify sensorimotor deficits with functional and anatomical neuroimaging techniques. More specifically, we will exploit experimental protocols that have been developed in basic research, that use a robotic exoskeleton coupled with a virtual reality device, to precisely quantify motor and proprioceptive deficits in stroke patients. Then, we will link these behavioral data to electroencephalographic (EEG) signals recorded during a motor adaptation task, as well as to anatomical data, namely conventional magnetic resonance imaging (MRI) completed by diffusion tensor images (DTI) in order to achieve a finer description of the cerebral lesions.

The present study will include two experimental parts, respectively centered on the proprioceptive deficits (Part 1) and the anomalies in the processing of movement-execution errors (Part 2).

Proprioceptive deficits in stroke patients : We will test the hypothesis that, when present, deficits in kinaesthesia and troubles in unconscious proprioception contribute substantially to motor deficits in stroke patients ; with as a corollary hypothesis, that deficits in " proprioception for action " are more determinant than deficits in the conscious sense of position (classically tested in clinics). In this purpose, we will collect three sets of behavioral data, in chronic stroke patients and healthy control participants, respectively intended to assess a) motor deficits, b) troubles in conscious sense of position, and c) deficits in "proprioception for action". To better document the neuronatomical substrates of these different types of deficits. In this purpose, we will link the obtained behavioral data with the results of detailed analyses of the lesions of the tested stroke patients.

Anomalies in the processing of movement-execution errors in stroke patients : We will assess movement-execution error processing in stroke patients, in order to test the idea that anomalies in error processing might contribute to motor deficits in stroke patients. In this purpose, we will record an electrophysiological correlate (ERP) of movement-error processing during a motor adaptation task. We will analyse the relation between the modulation of this ERP and motor performance. We will also examine the relation between these two sets of data (behavioral and electrophysiological) and the behavioral data collected during the first part of the study (Proprioceptive deficits). This will provide us with insight into the relationship between proprioceptive deficits and cinematic error processing. As in the first part of the study, we will link the observed electrophysiological and behavioral anomalies with the results of a detailed analysis of the anatomical lesions of the tested patients.

Inclusion Criteria:

• Stroke patients with ischaemia of the upper limb for less than 6 months
• Patients with sensorimotor deficits

Exclusion Criteria:

• Patients treated with botulinum toxin for less than 6 months
• Alcoholic patients
• Patients who had surgery on upper limb