Authors:
Patrick Schembri
;
Richard Anthony
and
Mariusz Pelc
Affiliation:
Department of Computing and Information Systems, University of Greenwich, Greenwich London and U.K.
Keyword(s):
Brain Computer Interface (BCI), Electroencephalography (EEG), Event-Related Potential, P300 Speller.
Related
Ontology
Subjects/Areas/Topics:
Applications
;
Biofeedback Technologies
;
Biomedical Devices for Computer Interaction
;
Biomedical Engineering
;
Biomedical Instruments and Devices
;
Biosignal Acquisition, Analysis and Processing
;
Brain-Computer Interfaces
;
Devices
;
Human-Computer Interaction
;
Methodologies and Methods
;
Pattern Recognition
;
Physiological Computing Systems
;
Physiology-Driven Computer Interaction
;
Software Engineering
Abstract:
In this paper we investigate the viability, practicability and efficacy of eliciting P300 responses based on the P300 speller BCI paradigm (oddball) and the xDAWN algorithm, with five healthy subjects; while using a non-invasive Brain Computer Interface (BCI) based on low fidelity electroencephalographic (EEG) equipment. The experiments were performed in three distinctive environments: lab conditions, mild and controlled user distractions, and real world environment (realistic sound and visual distractions present). Our main contribution is the assessment of the ways and extents to which different degrees of user distraction affect the detection success achievable using low fidelity equipment. Our results demonstrate the applicability of using off-the-shelf equipment as a means to successfully and effectively detect P300 responses, with different degrees of success across the three distinctive types of environment.