This is a randomised, within-subject, double-blinded, placebo-controlled study to assess the fMRI correlates of central nervous system (CNS) response to taVNS in early vs late-stage Parkinsons disease (PD), at two different frequencies. Forty PD patients will be included: twenty patients with early-stage (Hoehn and Yahr stage (HY) 1-2) and twenty patients with late-stage (HY 3-4) PD. Each patient will undergo one stimulation session, where they will receive active taVNS at 100 Hz (taVNS100), taVNS at 25Hz (taVNS25) or will have positioned electrodes without stimulation (xVNS). The order of taVNS100, taVNS25 and xVNS will be randomized between subjects. The session will include (1) xVNS and taVNS; (2) MRI and fMRI and (3) heart rate and breathing rate monitoring. Clinical assessment with the Movement Disorder Society-Unified Parkinson's Disease Rating Scale will be done before the session.

Each participant will complete 1 visit, where the investigators will first assess the MDS-UPDRS score, afther that the sesnsitivity threasholds for taVNS at 25Hz and taVNS at 100Hz will be assessed. The Nemos® electrode for taVNS will be positioned in the left cymba conchae with the following parameters: square-shaped pseudobiphasic pulse, interpulse duration 80μs, pulse width 300μs. taVNS25 will be applied at 25Hz and taVNS100 will be applied at 100Hz. The sensory threshold will be assessed for each frequency and in the MRI scanner a pulse intensity of 0,1mA above the perceptual threshold will be applied.

After that the participant will be escorted to the MRI scanner where the MRI-safe electrode will be positioned in the left cymba conchae. The participant will be instructed to hold their head still during the complete experiment and that they will receive 3 different electrical stimulation types during 3 separate sessions. They will be further instructed that a black dot that will appear on the screen before each stimulation session and that they should fix their gaze on the dot. The instructions will be repeated before each BOLD MRI session.

MRI will be performed at the Center for Clinical physiology at the Medical faculty in Ljubljana (Center za klinično fiziologijo; CKF), using a 3T Philips Achieva 3TX dStream (Best, The Netherlands).

The investigators will acquire high-resolution T1-weighted and T2-weighted (0.7 x 0.7 x 0.7 mm^3) volumes of the brain. The area of brainstem and cerebellum will be scanned using a T1-weighted spectral pre-saturation with inversion recovery (SPIR) sequence to enhance the contrast of neuromelanin structures (and better delineate LC). The three structural scans will be interspaced with 3 time slots, containing three 8 minutes long block design blood oxygen level-dependent (BOLD MRI; fMRI) sessions (~1.75mm^3 isometric voxel). During each fMRI slot participants will receive a different type of stimulation (taVNS,25, taVNS100 and xVNS). While the participant will receive electrical stimulation during taVNS25 and taVNS100, there will be no current applied during xVNS. To reduce any carryover effect the investigators will randomise the order of the 3 fMRI sessions. The order of stimulation will not be revealed until the automated data and statistical analysis.

The timeline of the complete imaging session would be as follows: Survey scan - Slot 1 - T1 - Slot 2 - T2 - Slot 3 - neuromelanin sequence. The session should last up to 60 min. During this time the investigators will monitor heart rate and breathing rate remotely.

For the analysis T1-weighted images will be first segmented into grey, white and cerebrospinal fluid (CSF) tissue compartments using DARTEL, implemented in SPM12. This also provides spatial normalization and co-registration to the standard MNI space. Images from other modalities will first be co-registered to T1-weighted structural scans and next to the MNI template as secondary images using the identical transform algorithm. From the fMRI data the investigators will analyse (i) brainstem activation patterns (ROI analysis); and (ii) block design connectivity between nucleus tractus solitarii and central nervous system (CNS) structures and locus coeruleus and CNS stuctures. fMRI analyses will be done with Analysis of Functional NeuroImages (AFNI) tools.

Inclusion Criteria:

• diagnosis of Parkinson's disease according to the UK PD Society Brain Bank criteria
• ability to walk independently for at least 2 minutes
• cognitive impairment that might prevent cooperation during tests
• stable antiparkinsonian medications during the previous 1 month

Exclusion Criteria:

• medication resistant rest tremor or dyskinesias
• history of seizures
• implanted medical devices or metal implants
• pregnancy
• claustrophobia
• fixed or severe kyphosis
• any conditions that would prevent our patients from lying still for 1 hour in a supine position (i.e. muscle pain, degenerative joint diseases…).

The investigators plan to share the anonymised data with researchers upon request.

Upon request after study publication

Data will be shared with researchers upon reasonable request

Before the experiment the stimulation threshold (i.e. current) for each stimulation (taVNS25 and taVNS100) will be assessed by a third person.

The participant will be told, that the investigators are comparing the effect of 3 different stimulation types and that each will be delivered at a different intensity.

The order of stimulation will be randomised across the 3 fMRI slots by a third person and the stimulation type will be noted in a separate log file.

The outcomes assessor will be blinded to the stimulation type until the end of the analysis process.