The effect of brainwave-based feedback on anxiety and stress in college students
| ISRCTN | ISRCTN40165598 |
|---|---|
| DOI | https://doi.org/10.1186/ISRCTN40165598 |
| Secondary identifying numbers | PID 205866 |
- Submission date
- 14/10/2025
- Registration date
- 16/10/2025
- Last edited
- 15/10/2025
- Recruitment status
- No longer recruiting
- Overall study status
- Ongoing
- Condition category
- Mental and Behavioural Disorders
Plain English summary of protocol
Background and study aims
College students commonly face anxiety and stress issues, yet there is currently a lack of real-time, non-invasive monitoring and intervention methods. This study aims to explore, through two experiments, the capability of electroencephalogram (EEG) in predicting stress biomarkers (such as cortisol and alpha-amylase) and anxiety states, and to develop an EEG-based closed-loop neurofeedback training system to help individuals with anxiety regulate their emotions and alleviate symptoms.
Who can participate?
1. Anxiety Disorder Participants: College students aged 17-30 with current anxiety symptoms only, no history of other mental disorders, and no current use of psychiatric medications.
2. Healthy Participants: College students aged 17-30 with no history of mental or psychological disorders and no significant physical illnesses.
All participants must have normal vision and hearing (or corrected to normal) and sign an informed consent form.
What does the study involve?
Study 1: Participants will complete a modified version of the Montreal Stress Test, with the core task being mental arithmetic training. The experimental group will perform the task under time pressure, while the control group will have no time constraints. Electroencephalogram (EEG) data will be recorded simultaneously, and saliva samples will be collected for analysis of cortisol and alpha-amylase levels.
Study 2: Participants were randomly assigned to two groups: The experimental group received EEG-based closed-loop neurofeedback training to learn negative emotion suppression through real-time feedback during an emotional image recognition task. The control group performed the same task without receiving neurofeedback. All participants completed anxiety scale assessments and EEG recordings both before and after the intervention.
What are the potential benefits and risks of participation?
Potential Benefits:
1. Helps participants better understand and regulate their emotional states;
2. Accesses professional psychological assessments and EEG monitoring;
3. May alleviate anxiety symptoms and enhance emotional regulation skills.
Potential Risks:
1. Temporary discomfort or anxiety caused by task pressure during the experiment;
2. Mild discomfort from wearing EEG devices;
3. Some participants may experience fatigue or cognitive load during training.
Where is the study conducted?
Hefei Fourth People's Hospital (Anhui Mental Health Center) (China)
When does the study start and how long is it expected to last?
January 2022 to December 2027
Who is funding this research?
Anhui Province Key Research and Development Program Project (China)
Who is the main contact?
Wan Li, wanli@ahmu.edu.cn
Contact information
Public, Scientific, Principal Investigator
No.316 Huangshan Road, Shushan District, Hefei City, Anhui Province
Hefei
230026
China
| Phone | +86 (0)55163616273 |
|---|---|
| wanli@ahmu.edu.cn |
Public
No.316 Huangshan Road, Shushan District, Hefei City, Anhui Province
Hefei
230026
China
 ORCID ID |
0009-0003-8986-9035 |
|---|---|
| Phone | +86 18855183148 |
| 2230577076@qq.com |
Study information
| Study design | Interventional randomized controlled trial |
|---|---|
| Primary study design | Interventional |
| Secondary study design | Randomised controlled trial |
| Study setting(s) | University/medical school/dental school |
| Study type | Treatment, Efficacy |
| Participant information sheet | Not available in web format, please use contact details to request a participant information sheet. |
| Scientific title | A randomised controlled trial of a closed-loop EEG neurofeedback intervention for anxiety: targeting emotion regulation and event-related potential biomarkers |
| Study objectives | Experiment 1 aims to explore the predictive capability of EEG feature models for cortisol and α-amylase levels by integrating different machine learning models and feature importance analysis methods, with the goal of providing novel solutions for real-time, non-invasive stress monitoring. Experiment 2 aims to focus on specific ERP components during states of anxiety and to explore a brainwave feature model capable of monitoring and predicting anxiety states (diagnostic model). |
| Ethics approval(s) |
Approved 05/01/2023, Medical Research Ethics Committee of Hefei Fourth People's Hospital (316 Huangshan Road, Shushan District, Hefei, 230026, China; +86 (0)551 6361 6193; 2514331322@qq.com), ref: HFS-IRB-PJ-WL (2022002) |
| Health condition(s) or problem(s) studied | Anxiety disorder |
| Intervention | Study 1: The participants underwent two identical experimental procedures: Day 1: the first montreal imaging stress task (MIST) with EEG recording. Day 4: the second MIST task with EEG recording. The MIST task included four 3-minute blocks: the first and third blocks were control blocks and the second and forth blocks were experimental blocks , with saliva collection completion after each block. A final saliva collection was completed after a 15-minute rest. Saliva was collected six times: after MIST training (following a 5-minute rest) and at the end of each 3-minute block (four times), with a final collection after a 15-minute rest. A total of 12 saliva samples were collected before and after the intervention (Figure 1). Saliva samples were collected via standardized time points, cryopreserved at -80°C, and tested for cortisol and alpha-amylase levels using kits. Study 2: All participants were randomly assigned to one of the following two groups: Experimental group (n=30): Participants exhibiting anxiety symptoms receive closed-loop neurofeedback training based on electroencephalogram (EEG). During training, participants performed an emotion image recognition task (positive/negative) while actively suppressing negative emotions triggered by viewing negative images. Task-related EEG signals were collected in real-time, and real-time neural feedback was provided based on event-related potential components (e.g., N2/P3/LPP). Each training session lasted approximately 30 minutes, with multiple sessions conducted (specific number determined by experimental design). Control group (n=30): Received identical intervention content to the experimental group but without neural feedback. Participants performed the same emotion image recognition task while EEG data was collected. All participants completed clinical scale assessments (e.g., Self-Rating Anxiety Scale) and baseline EEG recordings prior to intervention. EEG data collection and scale assessments were repeated post-intervention. |
| Intervention type | Procedure/Surgery |
| Primary outcome measure | Study 1: Prediction of cortisol and α-amylase levels using machine learning models based on electroencephalogram features. Study 2: 1. Change in amplitude of anxiety-specific Event-Related Potential (ERP) components (e.g., N2, P3, or Late Positive Potential (LPP)) during the emotional regulation task. Timepoint: From baseline (pre-intervention) to immediately after the completion of the closed-loop training sessions. 2. Change in high-frequency beta wave power during the emotional regulation task. Timepoint: From baseline (pre-intervention) to immediately after the completion of the closed-loop training sessions. |
| Secondary outcome measures | Study 1: 1. Performance comparison of different machine learning models in predictive tasks: Specific metrics include accuracy, which evaluates the performance of multilayer perceptrons, support vector machines, and linear discriminant analysis in predicting elevated/low cortisol and α-amylase levels. 2. Importance contribution of EEG features to predictive models: Quantification of relative contributions from different frequency bands (e.g., Delta, Theta, Alpha, Beta, Gamma) to cortisol and α-amylase prediction using SHAP values. Quantification of relative contributions from different brain regions (e.g., right prefrontal cortex, occipital cortex, parietal cortex) to predicting these biomarker levels through SHAP values. 3. Regression prediction accuracy of machine learning models for cortisol and α-amylase levels: Evaluation metrics include R² (coefficient of determination) and mean absolute error to measure models' ability to predict continuous physiological parameter values. 4. Differences in EEG feature model prediction effectiveness across time points (first and second MIST tasks): Assess whether temporal factors during stress task interventions affect the classification performance of EEG feature-based models. Study 2: 1. Change in self-reported anxiety scores as measured by standardized anxiety scales (e.g., GAD-7, SAS). Timepoint: From baseline (pre-intervention) to immediately after the completion of the training sessions. 2. Change in behavioral performance during the emotional regulation task, including: Reaction time to positive and negative stimuli. Accuracy of emotion identification. Timepoint: From baseline to post-intervention. 3.Change in scores from the closed-loop training task (where a higher score indicates better ability to suppress negative emotions and greater emotional stability). Timepoint: Measured after each training block and compared across the training course. 4.Changes in other EEG/ERP components not listed as primary outcomes, such as P1, N1, and Error-Related Negativity (ERN/ΔERN), and their correlation with anxiety reduction. Timepoint: From baseline to post-intervention. 5.Participant's subjective experience and strategy use, assessed through post-training questionnaires or interviews, regarding: 5.1. Attention concentration level. 5.2. Cognitive load. 5.3. Emotional state. 5.4. Fatigue level. 5.5. Self-regulation strategies developed. Timepoint: After each training session or at the end of the entire intervention. 6.Performance and generalizability of the predictive EEG-based anxiety diagnostic model, evaluated by: The model's accuracy, sensitivity, and specificity in classifying or predicting anxiety states. Timepoint: Upon completion of data collection and analysis. |
| Overall study start date | 07/08/2022 |
| Completion date | 31/12/2027 |
Eligibility
| Participant type(s) | Learner/student |
|---|---|
| Age group | Adult |
| Lower age limit | 17 Years |
| Upper age limit | 30 Years |
| Sex | Both |
| Target number of participants | 60 |
| Key inclusion criteria | Anxiety disorder participants: 1. Age 17-30 years old college students; 2. Able to comprehend questionnaire content and complete all scales independently; 3. Currently experiencing only anxiety symptoms without other DSM-5-confirmed mental disorders; 4. No history of psychiatric medication use (within 6 months); 5. Normal vision and hearing or corrected vision/hearing; 6. No severe physical illnesses, no history of head trauma, no metal implants in the head, no epilepsy, no episodes of transient loss of consciousness; willing to participate and have signed an informed consent form. Healthy participants: 1. Age 17-30 years old college students 2. Normal hearing, normal or corrected vision 3. Right-handed 4. No previous diagnosis of mental or psychological illness 5. No history of major physical illness |
| Key exclusion criteria | Exclusion criteria for anxiety disorders: 1. The questionnaire results contain significant logical inconsistencies; 2. Confirmed diagnosis of mental illness or family history of psychiatric disorders; 3. History of head trauma or implanted devices in the head; 4. If participants show marked improvement in anxiety levels or nearly perfect scores during closed-loop training before the assessment, they may be preliminarily excluded from the anxiety group as they may not meet the diagnostic criteria for anxiety. |
| Date of first enrolment | 02/03/2023 |
| Date of final enrolment | 15/04/2024 |
Locations
Countries of recruitment
- China
Study participating centre
Shushan District
Hefei
230032
China
Sponsor information
Hospital/treatment centre
No.316 Huangshan Road, Shushan District, Hefei City, Anhui Province
Hefei
230026
China
| Phone | +86 18855183148 |
|---|---|
| wanli@ahmu.edu.cn |
Funders
Funder type
Government
No information available
Results and Publications
| Intention to publish date | |
|---|---|
| Individual participant data (IPD) Intention to share | Yes |
| IPD sharing plan summary | Available on request |
| Publication and dissemination plan | Planned publication in a peer-reviewed journal |
| IPD sharing plan | The datasets generated during and/or analysed during the current study will be available upon request from Wan Li (wanli@ahmu.edu.cn) after the publication of the article |
Editorial Notes
15/10/2025: Trial's existence confirmed by Medical Research Ethics Committee of Hefei Fourth People's Hospital.
