Type: Bachelor-/ Master Thesis
Online Recognition of Error-related Potentials in Brain Computer Interface systems
Brain Computer Interfaces (BCIs) allow interaction between humans and machines by decoding brain signals into user intentions. Incorrect recognition of intentions decreases the overall channel bitrate, which is inherently low. Key to efficient communication is the ability to mitigate these errors. From the literature, it is known that error-related potentials are elicited in the brain, indicating human awareness of erroneous responses done by him-/herself or by observed humans or robots.
This thesis aims at investigating the existence of a special type of those potentials in non-invasive electroencephalographic BCI (EEG-BCI), namely the interaction error-related potentials. The basic setup of this work will be built around a navigational maze puzzle game, as a way to stimulate these potentials, and a way to learn the classifier.
The main goal here is to recognize these error potentials, rapidly and accurately. Additionally, navigation in the maze should be carried out with already implemented BCI paradigms (i.e. SSVEP).
Tasks:
- Literature review of error-related potentials and algorithms for online recognition
- Design of a maze puzzle game
- Online classification of interaction error-related potentials
- Usage of available BCI paradigms (i.e. SSVEP) as means of communication/control
Supervisor:
Literature:
- P. Ferrez and J. del R. Millan. Error-related EEG potentials generated during simulated brain-computer interaction . In IEEE transactions on biomedical engineering, vol. 55, no. 3, pp. 923-9, 2008.