PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Data Driven Neuroergonomic Optimization of BCI Stimuli
Michael Tangermann, Johannes Höhne, Martijn Schreuder, Max Sagebaum, Benjamin Blankertz, Andrew Ramsay and Roderick Murray-Smith
In: 5th International BCI Conference, 22-24 Sept 2011, Graz.

Abstract

Neuroergonomic design of Brain-Computer Interface (BCI) experiments can be realized as a data driven optimization of stimuli. The goal of this process is to increase the number and information content of class-discriminant features of the EEG for the BCI task at hand. While existing electrophysiological literature indicated the influence of confounding variables on e.g. P300 latency and amplitude by group studies and grand average statistics, the BCI performance can be boosted by a large amount when optimized stimuli are explored and designed for individual users and then used in single-trials. The potential of this design principle is shown in an offline analysis for the example of a visual (n = 8) and an auditory (n = 5) ERP study with healthy subjects, where the optimization of stimulus parameters leads to both a decrease in classification errors and an increase in speed.

PDF - Requires Adobe Acrobat Reader or other PDF viewer.
EPrint Type:Conference or Workshop Item (Paper)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Brain Computer Interfaces
Multimodal Integration
ID Code:9452
Deposited By:Michael Tangermann
Deposited On:16 March 2012