Talk by Dr. Fabien Lotte (Inria Bordeaux Sud-Ouest, France)

Speaker:
Fabien LOTTE, PhD
Inria Research Scientist
Inria Bordeaux Sud-Ouest/LaBRI, France
team Potioc - http://team.inria.fr/potioc

Title:
Combining machine learning and psychology to design usable Brain-Computer Interfaces

Abstract:
Brain-Computer Interfaces (BCIs) are systems that can translate brain activity patterns of a user into messages or commands for an interactive application. Such brain activity is typically measured using Electroencephalography (EEG), before being processed and classified by the system. EEG-based BCIs have proven promising for a wide range of applications ranging from communication and control for motor impaired users, to gaming targeted at the general public, real-time mental state monitoring and stroke rehabilitation, to name a few. Despite this promising potential, BCIs are still scarcely used outside laboratories for practical applications. The main reason preventing EEG-based BCIs from being widely used is arguably their poor usability, which is notably due to their low robustness and reliability. There is thus a need to make BCI more usable.

In our research, we aim at making BCI usable by combining tools and methods from both machine learning and psychology. This talk will illustrate our current research and open research questions in this area. First, I will discuss how BCI can be made more usable by studying and improving the BCI user training, to ensure that users can learn to control a BCI efficiently and effectively. Here psychology is used to borrow principles from educational psychology and human learning theories to guide user training, while machine learning is used to model such training computationally, as well as to design user specific feedback. Then, I will describe how BCI could be used for Neuroergonomics, here is for estimating the ergonomics qualities of human computer interfaces using EEG. In that area we use methods from psychology to induce and study specific cognitive states relevant for ergonomics, e.g., mental workload. Machine learning is then used to recognize those mental states in EEG signals. We will conclude by some open research directions in machine learning for the BCI field.