Conventional rTMS is applied to the left dorsolateral prefrontal cortex (DLPFC) using the standard 5 cm rule for scalp-based localization. Stimulation is delivered with intermittent theta-burst stimulation (iTBS) at 90% resting motor threshold (RMT), 3 pulses at 50 Hz per burst, repeated at 5 Hz, for a total of 1800 pulses per session. Each session lasts about 570 seconds. Participants complete 20 sessions over 4 weeks, with 2 sessions per day on 2 days per week.

Sham rTMS is delivered using a sham coil or low-intensity stimulation (10% RMT) to mimic the clicking sounds and scalp sensations of active rTMS without producing cortical activation. The sham sessions are matched to the active treatment in frequency, duration, and total number: 20 sessions in 4 weeks (2 sessions/day, 2 days/week, 570 seconds per session).

G-FCQ-T The General Food Craving Questionnaire-Trait (G-FCQ-T) is a self-reported instrument used to assess the general tendency to experience food cravings. It comprises 21 items. Participants rate each item on a 6-point Likert scale ranging from 1 ("Never") to 6 ("Always"). The total score is the sum of all item ratings, with a possible range from 21 to 126. A higher total score indicates a more severe trait propensity for food craving.

G-FCQ-S The General Food Craving Questionnaire-State (G-FCQ-S) is a self-reported instrument used to assess the current, momentary state of food craving. It comprises 15 items. Participants rate each item on a 6-point Likert scale ranging from 1 ("Strongly Disagree") to 6 ("Strongly Agree"). The total score is the sum of all item ratings, with a possible range from 15 to 90. A higher total score indicates a more intense current state of food craving.

Subjective food craving is assessed using a Visual Analogue Scale (VAS). Participants indicate their current, momentary level of food craving by marking a point on a 100-mm horizontal line. The left endpoint (0 mm) is labeled "No food craving at all," and the right endpoint (100 mm) is labeled "The most intense food craving imaginable." The score is determined by measuring the distance from the left endpoint to the participant's mark, ranging from 0 to 100. A higher score indicates a more intense subjective feeling of food craving.

The Alcohol Use Disorders Identification Test (AUDIT) is a simple, self-reported screening tool developed by the World Health Organization (WHO) to identify hazardous and harmful patterns of alcohol consumption. The questionnaire consists of 10 items. The first 8 items are scored on a scale from 0 to 4. Items 9 and 10 are scored 0, 2, or 4. The total score is the sum of all item ratings, with a possible range from 0 to 40. A higher total score indicates a greater likelihood of hazardous and harmful alcohol use, as well as possible alcohol dependence.

The Penn Alcohol Craving Scale (PACS) is a self-reported questionnaire designed to measure the intensity of an individual's craving for alcohol over the past week. The scale consists of 5 items assessing the frequency, duration, and intensity of craving, the ability to resist drinking, and the overall rating of craving. Each item is rated on a 6-point scale from 0 to 6, with varying verbal anchors for each item (e.g., "Never" to "Nearly All the Time"). The total score is the sum of all 5 item ratings, with a possible range from 0 to 30. A higher total score indicates a more severe level of subjective alcohol craving.

Alcohol consumption is assessed using the Timeline Followback (TLFB) method, a structured interview that provides a quantitative estimate of an individual's daily drinking. Using a calendar to aid recall, participants retrospectively report their daily alcohol consumption over a specified period (e.g., the past 30 days). The data is used to calculate key metrics such as the number of drinking days, the number of heavy drinking days, and the average number of standard drinks consumed per day or per week. A higher value indicates a greater level of alcohol consumption.

The amplitude of low-frequency fluctuations (ALFF) will be calculated from resting-state fMRI data. It measures the intensity of spontaneous neural activity in specific brain regions. The value is typically presented as a normalized Z-score. A higher ALFF value indicates greater regional spontaneous neural activity.