Imagine steering your MarioKart racer using nothing but your thoughts. Neuralink has already demonstrated that brain signals can be used to control video games, as seen in this video: https://www.youtube.com/watch?v=F7am9DB0qq4.

But can we achieve something similar with wearable EEG headphones—systems that are much more practical but come with lower signal quality?

In this thesis, you will explore whether brain-computer interfaces (BCIs) based on consumer-grade EEG can provide meaningful control over a game like MarioKart. This challenge involves both experimental and AI hurdles, from designing a reliable experiment to training machine learning models that translate brain activity into in-game actions.

Possible directions for your project include (but are not limited to):

• Exploring the feasibility of steady-state visual evoked potentials (SSVEP) or motor imagery BCI paradigm for control – this will involve reading and testing
• Comparing different machine learning approaches for decoding EEG signals

Of course, testing your BCI will require a lot of time playing MarioKart—but all in the name of science!

Lastly, this project is a collaboration with researchers at the Politecnico di Milano. That means visits to Milan may be part of the project.

If you have any questions about the topic beforehand, please contact Michael Knierim (michael.knierim∂kit.edu).