Tourette syndrome (TS) is a complex neurodevelopmental disorder characterized by the occurrence of involuntary movements (motor tics) and vocalizations (phonic tics). The onset of TS is usually in childhood, and the prevalence of TS is estimated between 0.3 and 0.9% before the age of 18, decreasing progressively after that age. Most patients also suffer from associated psychiatric comorbidities (ADHD, OCD, mood disorders). Although the cause of TS remains unknown, the preferred hypothesis is the interaction of predisposing genetic factors and precipitating environmental factors (perinatal accidents, infectious diseases). From a pathophysiological point of view, it is widely demonstrated by structural, electrophysiological studies, functional neuroimaging, as well as by different animal models, that dysfunctions of the cortico-striato-pallido-thalamo-cortical loops (responsible for the regulation of movements, cognitive processes, and emotions) play a major role in the genesis of tics. Deep brain stimulation (DBS) treatment can be proposed as an invasive therapy in patients with severe TS resistant to usual treatments (psychotherapy, pharmacological treatments). In a well-selected population of drug-resistant patients, DBS allows an estimated overall improvement of 30 to 50% in the YGTSS score. The deep brain stimulation method currently used in TS is based on continuous (24/7) and undifferentiated stimulation (fixed electrical intensity). This stimulation paradigm, devoid of adaptability to the patient's symptoms, could be at the origin of undesirable effects (related to the modulation of physiological signals), of a sub-optimal efficiency, or of an unnecessary overuse of the stimulator's capacities (battery depletion). The development of new deep brain stimulation paradigms (closed-loop stimulation), allowing the identification of pathological neuronal activity and the dynamic adaptation of stimulation parameters to these neuronal signals, requires reliable and reproducible pathological biomarker, correlated with the occurrence of tics. However, in TS, electrophysiological abnormalities are still not well characterized, and most of the work published on the subject were based on intraoperative recordings and needs to be confirmed on recordings at a distance from the surgery before its potential use in closed-loop stimulation paradigms. Indeed, during the first weeks after surgery, different factors tend to modify the electrophysiological signals. Several questions arise at the end of this healing period: * Are these pathological oscillations (distinct from the brain oscillations induced by physiological voluntary movement) still detectable weeks after the surgery? * What are the temporal dynamics of these oscillations around a tic? * What is the spatial topography of these oscillations within the GPi? * Is there a strong inter-individual variability? * How are changes in cortical activity associated with these subcortical oscillations? * Are the modulations of pallidal activity alone sufficient to predict the occurrence of a tic? Thus, our study aims to define precisely the cortico-subcortical activity concomitant with the occurrence of a tic, and to identify reliable and reproducible biomarker(s) associated with tics in TS. In order to specify these biomarker(s), their temporal correlation to tic occurrence, their spatial distribution, as well as the dynamics and cortico-subcortical coherence of the identified abnormalities, we propose a prospective study on 10 patients with severe and drug-resistant TS, treated by bi-pallidal deep brain stimulation as part of routine care (no device implantation as part of the research). An evaluation of pallidal LFP synchronized with a high-resolution video-electroencephalography recording (128 to 256 sensors) will be performed at a distance (M+\[3-48\]) from surgery, in order to determine the variations in pallidal and electroencephalographic activity surrounding the occurrence of tics. A control condition with voluntary (tic-like) movement will be carried out in a second time, to distinguish the modifications related to the voluntary movement from those related to the occurrence of a tic. A reconstruction of the electrode positioning will be performed using the LeadDBS pipeline, and individual and group analyses will be performed to specify the mapping of pathological oscillations within the pallidum and throughout the cerebral cortex.