Presented at
30C3 (2013),
Dec. 29, 2013, 4 p.m.
(60 minutes).
The brain can be understood as a highly specialized information processing device. Because computers basically do the same thing, it's not too absurd to try to link these two together. The result is a brain-computer-interface.
This talk explains the core functionality of our brain and how to access the stored data from the outside. Software and hardware have already reached a somewhat hacker-friendly state, and we want to show you how we got there. We're also here to answer all your questions about the brain.
Communication between humans and computers has great tradition, but also underlies several disadvantages. Interfaces like mice, keyboards or microphones essentially link the user's body parts to a computer. When creating new content, these interfaces prove to be relatively inefficient, inaccurate and limited by the user's skill.
A good brain-computer-interface would take body movements out of the process. Less required skill and more information flow density are only the most obvious benefits. It is a potential replacement for dozens of today's specialized devices. And, just as a microphone does for voice, it would also allow the transfer of visual imaginations.
Before we explain how to create the most advanced brain-computer-interface possible today, Dominic dives quite deeply into the signal processing structure in the brain. He presents the most recent findings in cognitive science and explains what happens - step for step - when we imagine the image of a red ball. At the end of this first section, he arrives at the level of electrical signals.
Anne then takes over with the groundwork necessary to capture and process these electrical signals. She isn't afraid to use proper math for a deeper understanding, but she made sure that her talk is easy to follow for non-tech majors too. The foundation for a contemporary brain-computer-interface consists of several core algorithms, which are used in lab settings today and by enthusiasts tomorrow. She calls out the pitfalls when dealing with signals from live brains, and covers the technical limitations with crunching the data.
We finish with a real-world perspective. The power of today's available hardware and software is still limited, but our understanding of informations inside and outside the brain has improved drastically. We've come a long way since electrode-level pattern matching, and we'd be excited to show you some examples of what's possible today.
Presenters:
-
Dominic
I'm a PhD student at the Max-Planck-Institute for Cognitive and Brain Sciences in Leipzig and explore the processing of language in the brain.
-
Anne
I am a a PHD student at the Institute of Geometry and Practical Mathematics of the RWTH Aachen. My topic is exploring algorithms of source localization of brain activity using EEG/MEG data.
Links:
Similar Presentations: