Home // COGNITIVE 2012, The Fourth International Conference on Advanced Cognitive Technologies and Applications // View article

Co-adaptivity in Unsupervised Adaptive Brain-Computer Interfacing: a Simulation Approach

Authors:
Martin Spüler
Wolfgang Rosenstiel
Martin Bogdan

Keywords: Brain-Computer Interface(BCI); unsupervised adaption; co-adaptivity

Abstract:
A Brain-Computer Interface (BCI) allows a user to control a computer by pure brain activity. Due to the nonstationarity of the recorded brain signals, the BCI performance tends to decrease over time. Recently, adaption of the BCI has been proposed as a means to counter non-stationarity and help to stabilize the BCI performance. Since most adaption methods for BCI are analysed in an offline setting, one important factor is not taken into account: that also the user is adapting. While online experiments take into account the adaptive user, a comparison of different classifiers with the same data is always biased towards the method that was used for feedback and thereby does not allow a proper evaluation of the classifier in a co-adaptive environment. To solve this problem, we propose a simulation approach that simulates an adapting BCI user and allows to test and compare different adaptive algorithms considering the co-adaptivity between BCI and user. With this approach we can also show, under which conditions an adaption of the BCI improves performance and when the adaptive BCI and the adaptive user hinder each other and lead to a decrease in BCI performance.

Pages: 115 to 121

Copyright: Copyright (c) IARIA, 2012

Publication date: July 22, 2012

Published in: conference

ISSN: 2308-4197

ISBN: 978-1-61208-218-9

Location: Nice, France

Dates: from July 22, 2012 to July 27, 2012