| Title:
|
A Gamified Neurofeedback System for Real-Time Attention Training Using Wearable EEG and Unity
|
| Author(s):
|
1
|
| Subject:
|
Brain–Computer Interfaces (BCIs)
|
|
|
|
Presenter Author:
|
|
| Name:
|
Sarah Asadi
|
|
|
|
Author:
|
|
| Name 1:
|
Sarah Asadi - Corresponding - Presenter
|
| Affiliation 1:
|
Biomedical Engineering Department, Amirkabir University of Technology
|
|
|
| Name 2:
|
Golnaz Baghdadi
|
| Affiliation 2:
|
Biomedical Engineering Department, Amirkabir University of Technology
|
|
|
| Name 3:
|
|
| Affiliation 3:
|
|
|
|
| Name 4:
|
|
| Affiliation 4:
|
|
|
|
| Name 5:
|
|
| Affiliation 5:
|
|
|
|
| Name 6:
|
|
| Affiliation 6:
|
|
|
|
| Name 7:
|
|
| Affiliation 7:
|
|
|
|
| Name 8:
|
|
| Affiliation 8:
|
|
|
|
| Name 9:
|
|
| Affiliation 9:
|
|
|
|
| Name 10:
|
|
| Affiliation 10:
|
|
|
|
Abstract
|
Original Research
|
|
introduction :
|
Wearable technology now allows researchers to monitor cognitive states outside traditional laboratory settings. At the same time, interactive digital environments require more practical methods for real-time mental assessment and feedback. This project aims to measure attention levels from brain signals using a Random Forest machine learning model.
|
|
Methods:
|
To maximize real-world applicability and computational efficiency, we recorded a subject-based EEG dataset and optimized a single-channel prefrontal (AF7) pipeline using the Muse S headset. We applied electrode gel to ensure stable signal acquisition, maintaining low electrode impedance and baseline drift under acceptable thresholds. To isolate genuine cortical activity, we implemented a strict micro-window artifact rejection protocol, discarding segments with physical movement to prevent the model from learning motor patterns instead of cognitive states. Signal features were extracted utilizing relative frequency band power and waveform complexity. To ensure statistical integrity, we trained this subject-based classifier using GroupKFold cross-validation, effectively eliminating temporal data leakage. To test real-time performance and provide interactive feedback, we developed a cognitive game called NeuroRise by building a UDP-based network framework to connect the model to the Unity game engine. In this setup, users complete a two-minute Stroop test to induce cognitive load while the model predicts their mental state. This output drives continuous visual feedback, where a virtual balloon rises as concentration increases and falls as it drops.
|
|
Results:
|
Preliminary results demonstrate the system's ability to successfully separate cognitive states, achieving a baseline classification accuracy of 65 percent in initial subject-based trials.
|
|
Conclusion:
|
Ultimately, this work provides a practical, low-cost setup for real-time cognitive monitoring and serves as a baseline for future neurofeedback systems designed to improve attention.
|
| Attachment:
|
|
| Keywords:
|
Electroencephalography, Brain-Computer Interfaces, Feature Extraction, Machine Learning, Neurofeedback, Gamification
|
| Key message:
|
|
| Do you want your abstract to be indexed, if this is selected as one of the top 100
articles?
|
