This two-part clinical study investigates the mechanisms and intervention effects related to emotional inhibitory control deficits in individuals with Alcohol Use Disorder (AUD), with a specific focus on brain-heart coupling (also referred to as brain-heart interplay, BHI). The study aims to identify the neurophysiological characteristics underlying inhibitory control impairments in AUD and evaluate whether non-invasive vagus nerve stimulation combined with mindfulness training can enhance emotional regulation and cognitive function.
Study 1: Mechanism Exploration Study 1 adopts a case-control observational design. A total of 28 individuals diagnosed with AUD (based on DSM-5 criteria) and 28 age- and education-matched healthy controls are recruited. Participants complete an emotional Go/NoGo task under EEG-ECG hyperscanning. Emotional inhibitory control performance is assessed through behavioral indices (reaction time, error rate) and further analyzed using the Hierarchical Drift Diffusion Model (HDDM) to quantify latent cognitive processing parameters (e.g., drift rate, boundary separation, non-decision time). Brain-heart coupling is evaluated through resting-state EEG-ECG data using multiscale coupling coefficients and maximal information coefficient (MIC) between heart rate variability (HRV) and cortical oscillations (particularly α, β, and θ bands).
Primary outcomes include:
Drift rate and inhibitory control accuracy under emotional interference Degree of BHI (HRV-EEG coupling strength and directionality) The moderating effect of self-reported emotion regulation ability (measured via DERS)
Study 2: Intervention Evaluation
Study 2 is a single-blind randomized controlled trial (RCT) involving 60 AUD participants randomly assigned to:
Active group (n=30): taVNS combined with mindfulness training Control group (n=30): sham stimulation combined with mindfulness training Participants in both groups receive daily 30-minute sessions for 10 consecutive days. taVNS is delivered using a wearable ear-clip device targeting the auricular branch of the vagus nerve (cymba conchae), with stimulation parameters set at 25 Hz, pulse width of 200 μs, and intensity adjusted to individual sensory threshold. Mindfulness training is conducted by a trained therapist, focusing on breath awareness and body scanning.
Outcome measures are collected pre- and post-intervention and include:
Emotional inhibitory control performance under the Go/NoGo paradigm HDDM-derived cognitive parameters (e.g., drift rate) EEG-ECG-based BHI metrics (HRV-EEG coupling during rest) Self-reported anxiety (GAD-7), depressive symptoms (PHQ-9), and alcohol craving Adverse events monitoring and adherence logs Data Acquisition and Processing EEG signals are recorded using a 64-channel portable system (10-20 system); ECG is simultaneously recorded.
Signals are preprocessed in EEGLAB (bandpass filtering, artifact correction via ASR), with HRV features extracted via Kubios.
HDDM modeling is conducted in Python (Bayesian MCMC estimation), and BHI is analyzed through mutual information, MIC, and transfer entropy metrics.
Sample Size and Statistical Plan Sample size was estimated based on prior effect sizes observed in similar taVNS and mindfulness interventions. Power analysis indicated 28 subjects per group achieves \>80% power to detect moderate group × time interaction effects.
Statistical analyses include:
Mixed-effects ANOVA for behavioral and neural outcomes Regression models to assess mediation and moderation effects (e.g., emotion regulation ability) Pearson/Spearman correlation between physiological and behavioral measures All analyses follow intention-to-treat principles with imputation for missing data where appropriate (e.g., multiple imputation for \<10% missing).
