Adding Human Learning in Brain--Computer Interfaces (BCIs): Towards a Practical Control Modality
summary
CLBCI implements single-trial classification on EEG motor imagery using FastICA, bandpass filtering, 1 s windows with 250 ms overlap and minimum-distance labeling, making the system operational in about one minute instead of 10–20. A Wachspress-based 2D polygon visualization shows per-class distance scores and lets users adjust a computational distance parameter as explicit feedback, triggering incremental learning. Tested on an event-based shooter game with four feedback conditions, the combined system+user feedback condition yields about a 10% absolute accuracy gain over baseline.
AI-written summary · about
chapters (9)
- 0:00Introduction to BCIs
- 1:36How brain signals work
- 3:26Active BCIs and training challenges
- 6:34Human learning in BCIs
- 9:01New system design with user feedback
- 11:50Visualization techniques for feedback
- 13:14Game-based validation results
- 15:00Conclusion and future work
- 15:28Q&A session
related talks
36mMachine-Brain Intelligence in Brain-Computer Interfaces
42mUnveiling the mysteries of Brain Computer Interfaces
09mNeuroAssist: Enhancing Cognition Computer Synergy with Adaptive AI & Enhance Neural Decoding
31mLearn to control your brain: Brain Computer Interfacing with Python
11mConnected brains
53mBrain: The ultimate hack
07mEnhancing In-game Immersion Using BCI-controlled Mechanics
08mHow to control drones with your mind: The Black Swan BCI headset