Disturbances in energy metabolism significantly increase the risk of developing major depressive disorder (MDD), especially in individuals with type 2 diabetes. Insulin sensitivity may particularly impair reward anticipation and motivational processes, contributing to anhedonia, a core symptom of depression. Preclinical and clinical studies highlight the vagus nerve as a critical pathway mediating metabolic signals between the body and the brain, influencing motivational and affective states. The present study aims to evaluate whether acute transcutaneous auricular vagus nerve stimulation (taVNS) improves motivation and mood and whether individual differences in insulin sensitivity modulate these improvements. The investigators plan to recruit 60 patients with MDD and 60 control participants matched for age, sex, and body mass index (BMI), covering a wide BMI range (up to 40 kg/m²) and insulin sensitivity (including patients with type 2 diabetes). Participants will undergo comprehensive metabolic assessments, behavioral testing of reward anticipation, motivation, consummation, and learning, and ecological momentary assessments (EMA) coupled with continuous glucose monitoring to assess real-world motivational behavior and glucose dynamics. Furthermore, participants will undergo two neuroimaging sessions, involving both task-free and task-based functional MRI, during concurrent taVNS or sham stimulation, implemented in a randomized, single-blinded, crossover design. This study hypothesizes that individuals with lower insulin sensitivity, particularly those with MDD and pronounced anhedonic symptoms, will show greater motivational and neural responsiveness to taVNS. H1A. Individuals with depression (vs. controls) and higher anhedonia show greater deficits in reward-related behavior and lower insulin sensitivity. H1B. Across all participants, reduced reward-related behavior and higher anhedonia are associated with lower insulin sensitivity. H2A. tVNS (vs. sham) increases motivation for rewards, brain responses to rewards, and body-brain interactions across participants. H2B. These tVNS-induced effects are particularly pronounced in individuals with depression and stronger anhedonia who show reductions in these domains. H3A. Greater tVNS-induced effects (behavioral, neural, body-brain) are associated with lower insulin sensitivity.
To assess where individual reward deficits manifest, participants will undergo an intake session that includes clinical interviews, a fasting blood draw, and a battery of reward tasks (Reward Rating, Effort Allocation, Taste Test, Go-Nogo-Learning). Changes in symptoms and glucose levels will be evaluated using a wearable glucose sensor and ecological momentary assessments (EMA) over 2 weeks. To assess insulin sensitivity, the investigators will perform an oral glucose tolerance test, with concurrent stimulation (tVNS vs. sham; i.e., two sessions, randomized). Finally, in two neuroimaging sessions, the investigators will assess the effect of acute tVNS (vs. sham; randomized) on motivation and stomach-brain coupling using concurrent functional magnetic resonance imaging (fMRI) and electrogastrography (EGG). Washout between tVNS/sham days will be a minimum of 2 days. Condition order will be randomized, and the design is single-blind (participant). To characterize our sample, the investigators will also collect information using standardized questionnaires assessing personality traits, eating behavior, psychiatric symptoms, and physical activity. To further characterize our sample metabolically, the investigators will also collect blood samples to determine metabolic parameters (e.g., acyl-ghrelin, des-acyl ghrelin, insulin, glucose, triglycerides, HDL, LDL), and participants can opt in to collect data for genetic analyses as part of a Biobank. These measures will be used to describe the sample and will be explored as predictors to explain inter-individual intervention effects. * Personality traits related to reward and motivation: Behavioral Inhibition/Activation System * Eating behavior: Three Factor Eating Questionnaire * Psychiatric symptoms: depressive symptoms and anhedonia (BDI-II; Snaith-Hamilton Pleasure Scale, German version, SHAPS-D; Temporal Experience of Pleasure Scale, TEPS; Dimensional Anhedonia Rating Scale, DARS). * Physical activity: International Physical Activity Questionnaire (IPAQ). During the EMA period, participants will measure changes in glucose using a continuous glucose monitor (CGM) using the Freestyle Libre 3 sensor. Participants will answer questions with respect to: * state ratings (e.g. hunger, mood) * anticipated rewarding activities (wanting, time planned) * consummated rewarding activities (liking, time spent) * as well as complete a food choice task.
Study Type
INTERVENTIONAL
Allocation
RANDOMIZED
Purpose
BASIC_SCIENCE
Masking
SINGLE
Enrollment
120
To stimulate vagal afferents, the electrode will be placed at the cymba conchae of the right ear using a previously established conventional stimulation protocol (30 s ON, 30s OFF, 25 Hz frequency, 250 µs pulse widths; tVNS R device, tVNS Technologies GmbH, Erlangen, Germany). Stimulation intensity will be pre-set for each participant for the following stimulation period to correspond to a mild pricking sensation determined with a staircase procedure in the lab session.
The control intervention consists of a sham stimulation. The electrode will be placed at the earlobe, which is not innervated by vagal afferent fibers. To improve blinding, the same stimulation protocol as for the tVNS will be applied (30 s ON, 30s OFF, 25 Hz frequency, 250 µs pulse widths; tVNS R device, tVNS Technologies GmbH, Erlangen, Germany) and stimulation intensities will be adjusted to correspond to a mild pricking sensation.
The oGTT will be conducted as a standardized metabolic challenge to assess glucose metabolism and insulin sensitivity. After an overnight fast, participants ingest a 75 g glucose solution. Venous blood samples are collected at five time points (0, 30, 60, 90, and 120 minutes) to measure plasma glucose and insulin concentrations. The primary variable of interest will be the peripheral insulin sensitivity index (ISI) according to Matsuda and DeFronzo as 10,000/(G0 × I0 × Gmean × Imean)1/2 with G = glucose and I = insulin. Additionally we will investigate the correspondence of ISI with other measures of insulin sensitivity (HOMA-IR), and Insulin secretion indices (Insulinogenic index (IGI), Corrected insulin response (CIR), Areas under the curve (AUC)), and HbA1c.
Section of Medical Psychology, Department of Psychiatry & Psychotherapy, Faculty of Medicine, University of Bonn
Bonn, Germany
RECRUITINGtVNS-induced behavioral changes in invigoration
The primary outcome of interest is the invigoration related to monetary and food rewards, which will be operationalized via the relative effort exerted on a grip force device using the effort allocation task. Invigoration is captured by the slope of the initial approach (i.e., the increase of force until a first plateau is reached). Invigoration will be compared between groups (CP, MDD) and stimulation conditions (tVNS, sham), and will be associated with insulin sensitivity. We will also investigate how effects on invigoration compare to effects on effort maintenance as well as on the association between ratings (i.e., wanting and exertion) during the effort task from the Intake session and effort (i.e., their correspondence expressed as slopes).
Time frame: During MRI scan of effort allocation task (20 minutes), tVNS vs. sham session
tVNS-induced changes in brain responses
Brain activity (BOLD signal) in response to monetary and food rewards and during exerting effort on a grip force device will be analyzed by focusing on brain regions associated with the vagal afferent pathway. Brain activity will be compared between groups (CP, MDD) and stimulation conditions (tVNS, sham), and will be associated with insulin sensitivity.
Time frame: During MRI scan of effort allocation task (20 minutes), tVNS vs. sham session
Association of insulin sensitivity with anhedonia severity
Here, insulin sensitivity (Matsuda ISI) is considered a predictor variable, and the primary outcome is the association coefficient with anhedonia (SHAPS).
Time frame: oGTT (120 minutes)
Association of insulin sensitivity with motivation during an effort allocation task
Here, insulin sensitivity (Matsuda ISI) is considered a predictor variable, and the primary outcome is the association coefficient with invigoration, which will be operationalized via the relative effort (frequency button presses, controller) using the effort allocation task. Invigoration is captured by the slope of the initial approach (i.e., the increase of force until a first plateau is reached). Motivation will be compared between groups (CP, MDD), and in association with anhedonia, insulin-sensitivity. We will also investigate effects on effort maintenance and subjective ratings during the effort task (i.e., wanting and exertion) as well as the association between ratings and effort (slopes).
Time frame: ~40 minutes during Intake Session
tVNS-induced changes in peripheral Insulin Sensitivity (oGTT)
Here, stimulation condition is considered a predictor variable, and the primary outcome is the association coefficient with insulin sensitivity (Matsuda ISI). Interactions of stimulation condition with depression and anhedonia will be tested. We will also compare the association with associations of HOMA-IR and insulin secretion.
Time frame: oGTT (120 minutes), tVNS vs. sham sessions
tVNS-induced changes in stomach-brain coupling
Stomach-brain coupling will be assessed using phase-locking values (PLV) at baseline and during tVNS/sham stimulation concurrent to electrogastrogram (EGG) measurements in the MRI. PLV will be compared between groups (CP, MDD) and stimulation conditions (tVNS, sham) with a focus on core regions of the gastric network (postcentral gyrus, cingulate gyrus, precuneus, occipital cortex, fusiform gyrus, inferior frontal gyrus, inferior and superior parietal lobe, thalamus, and inferior cerebellum) and the vagal afferent pathway (nucleus of the solitary tract (NTS), ventral tegmental area (VTA), substantia nigra, hypothalamus, amygdala, putamen, caudate, nucleus accumbens, hippocampus, insula, ventromedial prefrontal cortex, lateral orbitofrontal cortex, dorsal anterior cingulate cortex). PLV will be correlated with depressive symptoms, anhedonia, and insulin sensitivity.
Time frame: During MRI scan (up to 120 minutes)
tVNS-induced changes in gastric motility
The gastric myoelectric frequency will be assessed during the MRI sessions at baseline and during tVNS and sham stimulation using an EGG. Acute changes in gastric peak frequency will be compared between groups (CP, MDD) and stimulation conditions (tVNS, sham).
Time frame: During MRI scan (up to 120 minutes)
tVNS-induced changes in positive and negative affect
Mood will be assessed repeatedly before and after stimulation via visual analogue ratings (0-100). Mood will be surveyed by using items of the Positive and Negative Affect Schedule (PANAS). Changes will be compared between groups (CP, MDD) and stimulation conditions (tVNS, sham), and will be associated with insulin-sensitivity and anhedonia. During the OGTT sessions, mood will be queried 5 times: T0 (before stimulation), T1-T4 (during stimulation) which will be modelled using a 2-level factor timepoint (pre vs. post). Post-hoc analyses may include separate comparisons of T1-T4 to T0. During Neuroimaging days mood will be queried before and after stimulation and be modelled using a 2-level factor timepoint (pre vs. post). In addition, we will also evaluate mood ratings from the EMA data.
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.
Time frame: Before and after tVNS (vs. sham) stimulation (for oGTT, at the end of a 120 minutes stimulation period; for neuroimaging at the end of a 40 minutes stimulation period).
Association of insulin sensitivity with anticipation of daily rewarding activities
Here, insulin sensitivity (Matsuda ISI) is considered a predictor variable, and the primary outcome is the association coefficient with reward-related behavior. Rewarding activities will be queried during a \~2 week ecological momentary asseessment (EMA) period using smartphone questionnaires 2x daily. Anticipation of usually rewarding activities across several domains (hobbies, social, sensory, work, and food/drinks) will be measured using the average wanting ratings (0-100) and planned time (minutes). Anticipation will be compared between groups (CP, MDD), and in association with anhedonia, insulin-sensitivity. Fluctuations will be correlated with glucose levels from continuous glucose monitoring (CGM).
Time frame: During 2 week period, 2 times daily.
Association of insulin sensitivity with consummation of daily rewarding activities
Here, insulin sensitivity (Matsuda ISI) is considered a predictor variable, and the primary outcome is the association coefficient with reward-related behavior. Rewarding activities will be queried during a \~2 week ecological momentary asseessment (EMA) period using smartphone questionnaires 2x daily. Consummation of usually rewarding activities across several domains (hobbies, social, sensory, work, and food/drinks) will be measured using the average liking ratings (-100 to 100) and time spent with rewarding activities (minutes). Consummation will be compared between groups (CP, MDD), and in association with anhedonia, insulin-sensitivity. Fluctuations will be correlated with glucose levels from continuous glucose monitoring (CGM).
Time frame: During 2 week period, 2 times daily.
Association of insulin sensitivity with anticipatory reward ratings
Here, insulin sensitivity (Matsuda ISI) is considered a predictor variable, and the primary outcome is the association coefficient with anticipatory ratings for rewards. Participants will rate different reward types (food, erotic, music) regarding liking and wanting during the intake session (\~20 min). Ratings will be tested between groups (CP vs. MDD), in association with anhedonia, and insulin-sensitivity, as well as their interaction with reward type.
Time frame: ~40 minutes task during Intake Session
Association of insulin sensitivity with consummatory reward responses
Here, insulin sensitivity (Matsuda ISI) is considered a predictor variable, and the primary outcome is the association coefficient with consummatory reward ratings. During the Intake Session, participants will repeatedly rate food rewards regarding liking, wanting, and sensory qualities, gradually moving from anticipation to consummation (phase). Ratings will be compared between groups (CP, MDD) and associated with anhedonia and insulin sensitivity, and their interaction with phase.
Time frame: ~35 minutes during Intake Session
Association of insulin sensitivity with learning during a Go-noGo-learning Task
Here, insulin sensitivity (Matsuda ISI) is considered a predictor variable, and the primary outcome is the association coefficient with the learning rate. During Intake, participants will perform a go-no-go learning task. Learning rate will be tested between groups (CPs, MDD), and in association with anhedonia and insulin sensitivitiy. We will also investigate changes in "go bias" and "Pavlovian bias".
Time frame: ~20 minutes during Intake
Association of insulin sensitivity with daily choices between food rewards
Here, insulin sensitivity (Matsuda ISI) is considered a predictor variable, and the primary outcome is the association coefficient with reward-related behavior. During the EMA period, participants will repeatedly (\~2x daily, \~2 weeks) choose between two food rewards (which they have rated during the Intake similarly regarding liking). Choices will be compared between groups (CP, MDD), and in association with anhedonia and insulin-sensitivity. Fluctuations will be correlated with glucose levels from continuous glucose monitoring (CGM).
Time frame: ~2 weeks, ~2 times daily EMA period
tVNS-induced effects in interstitial glucose
Participants will wear continuous glucose monitoring (CGM) devices for about 2 weeks that measure interstitial glucose levels, during which participants will undergo oGTT sessions with tVNS vs. sham to. To derive estimates of glucose control, we will use indices derived from the iglu toolbox.
Time frame: 120 minminutes during an OGTT, tVNS vs. sham
tVNS-induced changes in brain responses during task-free fMRI
To analyze stimulation-induced changes in task-free brain activity, participants' individual data will be modeled on the first-level including the onset and duration of the stimulation (i.e., we assume a 'tonic' constant effect across the complete 10 min of stimulation including ON and OFF phases of the stimulation). For the second-level analysis, individual contrast images (stimulation - baseline) will be entered in a full-factorial model in SPM including the within-subject factor stimulation condition (tVNS, sham) and the between-subject factor group (MDD, HCP). Stimulation order, sex, age, and BMI will be included as covariates. Associations with insulin sensitivity and anhedonia will be tested.
Time frame: 10 minutes baseline (pre-stimulation) and 10 minutes stimulation (tVNS vs. sham) during MRI scan; task-free (inscapes).
tVNS-induced changes in functional connectivity
We will test stimulation-induced changes in functional connectivity of brain regions of the vagal afferent pathway by conduction seed-based connectivity analyses. To this end, we will analyze seed-based functional connectivity and include group (MDD, HCP; independent between-subjects factor), stimulation condition (tVNS, sham; dependent within-subjects factor), and phase (baseline, stimulation; dependent within-subjects factor) as factors. We will also test associations with insulin sensitivity and anhedonia. Covariates will include age, sex, BMI, and session order.
Time frame: 10 minutes baseline (pre-stimulation) and 10 minutes stimulation (tVNS vs. sham) during task-free MRI.
tVNS-induced changes in effort maintenance
Description: Effort maintenance related to monetary and food rewards will be operationalized via the relative effort exerted on a grip force device. Effort maintenance will be compared between groups (CP, MDD) and stimulation conditions (tVNS, sham), and will be associated with insulin sensitivity. During the Intake, participants answered subjective ratings related to monetary and food rewards on a visual analogue scale during the effort task. Ratings of "wanting" and "exertion" will also be compared between groups (CP, MDD), and will be associated with insulin sensitivity and anhedonia. We will then test the associations between wanting ratings (from Intake) and effort (from neuroimaging with stimulation) and whether slopes differ between groups (CP, MDD) and stimulation conditions (tVNS, sham), and slopes will be associated with insulin sensitivity and anhedonia.
Time frame: During MRI scan of effort allocation task (20 minutes), tVNS vs. sham session
Association of insulin sensitivity with continuous glucose monitoring
Here, insulin sensitivity (Matsuda ISI) is the primary outcome and measures of glycaemic control derived from continuous glucose monitoring (CGM) are predictor variables. To estimate glycemic control from cgm data we will use the iglu toolbox.
Time frame: CGM data from EMA period (2 weeks)
Association of continuous glucose monitoring and hunger ratings as a measure of interoceptive accuracy
The primary outcome will be the association coefficient (interoceptive accuracy) of hunger ratings ("subjective") and concurrent glucose levels (CGM data; "objective").
Time frame: CGM data from EMA period (2 weeks)