Neuropsychiatric fluctuations are frequent and debilitating non-motor disorders in Parkinson's disease. Subthalamic stimulation is an effective treatment for motor and non-motor neuropsychiatric fluctuations in Parkinson's disease, which has also allowed to better characterize the electrophysiological activity of the subthalamic nucleus in patients with Parkinson's disease. Numerous studies have been carried out on electrophysiological activity correlated with motor disorders, showing that akinesia is correlated with an abnormal synchronization in the beta frequency band (13-30Hz) in the dorsal motor part of the subnucleus. thalamic. This abnormal synchronization is reduced when the akinesia improves with dopaminergic treatment. On the cognitive-behavioural non-motor side, electrophysiological studies are less numerous, but suggest an involvement of the subthalamic nucleus (with an abnormal synchronization in the lower frequency bands). Currently, the electrophysiological correlates of neuropsychiatric fluctuations are unknown. In addition, all studies on electrophysiological activity are carried out immediately after the intervention, and before the implantation of the pacemaker. It is unknown whether the electrophysiological correlates of motor and non-motor fluctuations change following their transient improvement observed in the setting of chronic stimulation of the subthalamic nucleus and chronic reduction of dopaminergic treatment. The new PERCEPT stimulators (Medtronic) available on the market make it possible to record long-term electrophysiological activity for the first time. With this study we want to define the electrophysiological markers of neuropsychiatric fluctuations and their change over time.

In PD several studies have shown an abnormal synchronized activity in the beta band (13-30 Hz) in the motor dorsal STN, which correlates with bradykinesia. This abnormal activity is decreased by dopaminergic treatment and movement. In contrast, higher alpha activity is concentrated in the ventromedial part of the nucleus. Anatomically, it has been shown that the motor dorsal subthalamic nucleus is connected with the motor frontal cortex, whereas its non-motor ventral part is connected to the premotor and the prefrontal cortex.

More recently, an increasing attention has been focused on the key role of the STN in emotion and cognition in Parkinson's disease. Neuropsychiatric symptoms indeed represent a real burden for patients with PD and their caregivers. Patients might experience fatigue, depressive symptoms, anxiety, apathy, impulse control disorders, punding, hallucinations, and impaired cognition. Moreover, neuropsychiatric symptoms might fluctuate along the day. During the off state, patients might present with fatigue, lack of energy and motivation, low mood and slowness of thought, whereas during the on state they experience a sensation of well-being, high motivation, logorrhoea with easy flow of ideas, and good attention span and liveliness. Little is known about the electrophysiological correlates of such neuropsychiatric symptoms. An increase in the background activity and a reduced oscillatory activity in the alpha spectrum was observed in the ventral part of the right STN after vocal emotional stimuli. A change in the oscillatory activity in the alpha spectrum of the STN has been recorded after visual emotional stimuli. A theta activity in the ventral region of the STN has been recorded in PD patients with impulse control disorders while on medication.

Recently, a new tool, the neuropsychiatric fluctuations scale, has been developed. This scale, contrary to the other scales measuring the non-motor fluctuations, which are mainly retrospective, allows to measure the neuropsychiatric state of the patient at the very moment of the test. This is a critical advantage as patients can be anosognosic of their fluctuations. The neuropsychiatric fluctuations scale allows measuring the neuropsychiatric state of the patient "in real time", just as the MDS UPDRS assesses the motor condition.

At present, the electrophysiological marker of neuropsychiatric fluctuations in PD is unknown. Moreover, all electrophysiological data on STN activity comes from the postoperative phase and no data are available on their evolution over time.

The PERCEPT (Medtronic) allows to record STN activity not only in the immediate postoperative phase, but also in the medium and long-term follow-up, providing an excellent tool to investigate the electrophysiological activity of the subthalamic nucleus, and its changes over time.

In addition, while there are studies on the electrophysiological correlates of motor activity in PD, no studies are available on the change of this activity in the same patient after long follow-up. It is unknown whether the electrophysiological markers of motor and neuropsychiatric fluctuations change over time, together with the improvement of fluctuations induced by chronic stimulation and reduction of dopaminergic medication. From a clinical point of view, a motor and neuropsychiatric desensitization to levodopa has been shown in PD patients treated with chronic STN stimulation. Although relatively unknown and understudied, levodopa desensitization carries significant consequence in PD, such as malignant stimulation withdrawal syndrome. To determine how the electrophysiological markers of the OFF and the ON states change over time in the same patient not only will help to elucidate the mechanisms underlying this desensitization, but it will be also crucial for implementing algorithms for adaptive stimulation.

The purpose of our study is to record the STN activity in PD patients treated with STN stimulation during neuropsychiatric and motor fluctuations and their changes over time.

RESEARCH HYPOTHESIS AND EXPECTED RESULTS Our hypothesis is that neuropsychiatric fluctuations are associated with a change in the low-frequencies band activity in the ventral region of the STN, and that motor fluctuations are associated with changes in the beta band activity in the dorsal motor region of the STN. We expect a reduction of these abnormal activities over time, associated with the reduction of fluctuations.

Great efforts are to date made in order to create an adaptive stimulation, characterized by a real-time adaptation of parameters of stimulation to the fluctuations of the patient state. Adaptive stimulation would be more efficient than the current continuous stimulation system. Results of this study will allow to better understanding the pathophysiological basis of neuropsychiatric and motor fluctuation, and moreover to implement algorithms for create an adaptive stimulation.

Inclusion Criteria:

• Age between 30 and 75 years
• Diagnosis of Parkinson's disease with fluctuations and/or dyskinesias
• Ability to collaborate and fill a questionnaire in the OFF state condition (this criterion will be assessed based on the results of the preoperative screening to retain indication for STN stimulation, which takes place whenever a PD patient has to undergo DBS surgery)
• Surgery for bilateral STN stimulation with a Percept® (Medtronic) in the last 10 days
• Absence of dementia (defined as MoCA score > 24)

Exclusion Criteria:

• Active psychosis
• Patients unable to provide an informed consent or under gardianship
• Patients under judicial supervision or deprived of liberty
• Pregnancy and breastfeeding

PD patients with motor and non motor fluctuations, operated within the last 10 days for STN stimulation (PERCEPT implant)